Production Planning Software: When Does a Manufacturer Actually Need It?

1
00:00:00,000 --> 00:00:02,520
Here's the problem most manufacturers don't talk about.

2
00:00:02,520 --> 00:00:05,560
The planner already acts as the planning system most days.

3
00:00:05,560 --> 00:00:08,160
They sit between ERP dates, calls from the shop floor,

4
00:00:08,160 --> 00:00:09,720
a whiteboard, emails from sales,

5
00:00:09,720 --> 00:00:11,440
and that spreadsheet nobody else wants to touch

6
00:00:11,440 --> 00:00:13,920
because one broken formula can ruin the whole morning.

7
00:00:13,920 --> 00:00:16,200
Then a supplier calls, a machine goes down,

8
00:00:16,200 --> 00:00:19,080
and someone asks whether a rush order can still ship on Friday.

9
00:00:19,080 --> 00:00:20,680
So the planner becomes the connector.

10
00:00:20,680 --> 00:00:23,280
They check the order in ERP, call production to see

11
00:00:23,280 --> 00:00:25,480
what's actually running, look for material,

12
00:00:25,480 --> 00:00:28,000
and remember that one machine can only run the job

13
00:00:28,000 --> 00:00:31,280
after a tool change and only if the right operator is on shift.

14
00:00:31,280 --> 00:00:33,560
Then they move dates in Excel, send a message,

15
00:00:33,560 --> 00:00:35,360
and hope the new plan reaches everyone

16
00:00:35,360 --> 00:00:37,120
before the next disruption hits.

17
00:00:37,120 --> 00:00:38,600
That work takes real skill.

18
00:00:38,600 --> 00:00:41,440
It's not just admin, but it's a fragile way to run a factory

19
00:00:41,440 --> 00:00:43,360
when the number of orders constraints and changes

20
00:00:43,360 --> 00:00:44,280
keeps growing.

21
00:00:44,280 --> 00:00:46,160
You can have more data than ever.

22
00:00:46,160 --> 00:00:50,160
ERP, MES, machine data, power BI reports,

23
00:00:50,160 --> 00:00:52,080
and still have less confidence in the plan

24
00:00:52,080 --> 00:00:53,560
people are meant to follow.

25
00:00:53,560 --> 00:00:56,560
Because data doesn't automatically settle a planning decision,

26
00:00:56,560 --> 00:00:58,040
that brings us to the real question.

27
00:00:58,040 --> 00:01:00,680
And it's not, should we buy production planning software

28
00:01:00,680 --> 00:01:02,600
because other manufacturers have it?

29
00:01:02,600 --> 00:01:04,200
That's just software buying pressure,

30
00:01:04,200 --> 00:01:06,080
and it rarely leads anywhere useful.

31
00:01:06,080 --> 00:01:07,640
The better question is simpler.

32
00:01:07,640 --> 00:01:09,600
When has manual coordination stopped giving

33
00:01:09,600 --> 00:01:11,560
your people enough time and enough context

34
00:01:11,560 --> 00:01:13,000
to make a feasible plan?

35
00:01:13,000 --> 00:01:15,840
That's where we need to start, with the planning problem itself.

36
00:01:15,840 --> 00:01:18,800
Until you name the gap between what the planner holds together

37
00:01:18,800 --> 00:01:20,120
and what the system should handle,

38
00:01:20,120 --> 00:01:22,960
you're just buying tools for a problem you haven't defined.

39
00:01:22,960 --> 00:01:25,000
What production planning actually means?

40
00:01:25,000 --> 00:01:26,520
Production planning sounds like one thing,

41
00:01:26,520 --> 00:01:29,080
but it covers several decisions that happen at different levels

42
00:01:29,080 --> 00:01:30,360
and at different times.

43
00:01:30,360 --> 00:01:32,560
Broadly, planning asks whether the factory can

44
00:01:32,560 --> 00:01:35,040
meet expected demand with the people, machines,

45
00:01:35,040 --> 00:01:36,760
material, and time available.

46
00:01:36,760 --> 00:01:39,640
Closer to the shop floor, the question becomes much more direct,

47
00:01:39,640 --> 00:01:41,600
which order runs next on which resource,

48
00:01:41,600 --> 00:01:43,680
with which setup, and what does that choice delay?

49
00:01:43,680 --> 00:01:45,600
That's a different job from taking an order.

50
00:01:45,600 --> 00:01:47,480
Sales might promise a delivery date based on what

51
00:01:47,480 --> 00:01:48,480
seems possible.

52
00:01:48,480 --> 00:01:50,200
ERP might create a production order

53
00:01:50,200 --> 00:01:52,720
with a due date, a bill of material, and a routing.

54
00:01:52,720 --> 00:01:54,680
Both matter, but neither automatically

55
00:01:54,680 --> 00:01:56,800
creates an executable shop floor plan.

56
00:01:56,800 --> 00:01:59,640
An executable plan deals with the physical world.

57
00:01:59,640 --> 00:02:02,240
It knows that a work order needs a certain machine,

58
00:02:02,240 --> 00:02:04,480
maybe one of two machines, but one of them

59
00:02:04,480 --> 00:02:06,520
already carries a higher priority job.

60
00:02:06,520 --> 00:02:08,280
It knows the material might exist in stock,

61
00:02:08,280 --> 00:02:10,080
but still sit in incoming inspection.

62
00:02:10,080 --> 00:02:12,960
It knows that the machine can run for eight hours on paper,

63
00:02:12,960 --> 00:02:16,440
but a setup, preventive maintenance window, shift hand over,

64
00:02:16,440 --> 00:02:19,240
and missing operator cut that time down very quickly.

65
00:02:19,240 --> 00:02:21,640
Think of planning as a set of linked choices.

66
00:02:21,640 --> 00:02:25,480
Over the coming weeks or months, you decide what work should happen.

67
00:02:25,480 --> 00:02:28,000
Rough capacity checks, purchasing needs,

68
00:02:28,000 --> 00:02:30,480
subcontracting decisions, and a view of where

69
00:02:30,480 --> 00:02:33,240
future load could exceed plant capacity.

70
00:02:33,240 --> 00:02:35,520
After that, you move closer to execution.

71
00:02:35,520 --> 00:02:37,480
The horizon shrinks, the detail rises,

72
00:02:37,480 --> 00:02:39,320
and rough assumptions stop being enough.

73
00:02:39,320 --> 00:02:41,440
Now you need to decide the order of operations,

74
00:02:41,440 --> 00:02:44,120
accounting for actual resource calendars, current work

75
00:02:44,120 --> 00:02:46,800
in progress, setup sequences, material readiness,

76
00:02:46,800 --> 00:02:49,440
and any rule that limits where an operation can run.

77
00:02:49,440 --> 00:02:52,400
That detailed part is usually called production scheduling.

78
00:02:52,400 --> 00:02:55,720
Planning says we need to produce these orders within this period.

79
00:02:55,720 --> 00:02:58,520
Scheduling says this operation starts after that operation

80
00:02:58,520 --> 00:03:00,320
on this machine during this shift,

81
00:03:00,320 --> 00:03:03,560
because this is the least damaging feasible sequence.

82
00:03:03,560 --> 00:03:05,960
The word feasible matters more than fast.

83
00:03:05,960 --> 00:03:07,800
A schedule can look neat and still fail

84
00:03:07,800 --> 00:03:09,840
the moment production tries to follow it.

85
00:03:09,840 --> 00:03:12,240
If two operations claim the same machine at the same time,

86
00:03:12,240 --> 00:03:13,400
it isn't feasible.

87
00:03:13,400 --> 00:03:16,120
If the plan assumes material before receiving has released it,

88
00:03:16,120 --> 00:03:17,120
it isn't feasible.

89
00:03:17,120 --> 00:03:19,400
If it puts work on a resource that lacks the right tool

90
00:03:19,400 --> 00:03:22,120
or qualified operator, it's just a list of intentions.

91
00:03:22,120 --> 00:03:23,360
Let's make this concrete.

92
00:03:23,360 --> 00:03:25,440
Picture a normal Tuesday in a plant

93
00:03:25,440 --> 00:03:27,680
that produces several product variants.

94
00:03:27,680 --> 00:03:29,800
One order runs late on an upstream operation

95
00:03:29,800 --> 00:03:32,240
because the first batch fails a quality check.

96
00:03:32,240 --> 00:03:34,040
That order now needs extra time,

97
00:03:34,040 --> 00:03:35,880
and its next operation shares a machine

98
00:03:35,880 --> 00:03:38,080
with two other orders already in the queue.

99
00:03:38,080 --> 00:03:40,360
A manual plan might simply push the late order forward

100
00:03:40,360 --> 00:03:42,200
because its customer date looks urgent,

101
00:03:42,200 --> 00:03:43,120
but what does that change?

102
00:03:43,120 --> 00:03:45,400
Maybe the late order requires a long setup

103
00:03:45,400 --> 00:03:47,440
that disrupts the family of jobs already grouped

104
00:03:47,440 --> 00:03:48,800
to reduce changeovers.

105
00:03:48,800 --> 00:03:50,640
Maybe moving it ahead delays in order

106
00:03:50,640 --> 00:03:52,080
with a contractual ship date.

107
00:03:52,080 --> 00:03:54,600
Maybe a part needed for final assembly still hasn't arrived,

108
00:03:54,600 --> 00:03:57,560
so rushing the earlier operation doesn't improve delivery at all

109
00:03:57,560 --> 00:03:59,440
or maybe another approved route exists,

110
00:03:59,440 --> 00:04:01,600
but it uses a resource that runs near full load

111
00:04:01,600 --> 00:04:02,800
during the next shift.

112
00:04:02,800 --> 00:04:05,480
A real planning decision compares those consequences.

113
00:04:05,480 --> 00:04:08,200
It doesn't just move a coloured bar, change an ERP date,

114
00:04:08,200 --> 00:04:11,160
or tell the supervisor to fit it in somehow.

115
00:04:11,160 --> 00:04:13,600
And yes, fit it in somehow remains one of the more

116
00:04:13,600 --> 00:04:16,160
widely deployed planning algorithms in manufacturing.

117
00:04:16,160 --> 00:04:18,040
You also have to consider a time dimension

118
00:04:18,040 --> 00:04:19,760
that people often blur together,

119
00:04:19,760 --> 00:04:21,960
longer term planning works with uncertainty.

120
00:04:21,960 --> 00:04:24,360
Demand may change, suppliers may miss dates,

121
00:04:24,360 --> 00:04:26,840
and actual cycle times may differ from standards.

122
00:04:26,840 --> 00:04:29,080
You aren't trying to schedule every minute months in advance,

123
00:04:29,080 --> 00:04:31,280
you're trying to spot likely overloads early enough

124
00:04:31,280 --> 00:04:33,600
to take action, like adding a shift, moving work,

125
00:04:33,600 --> 00:04:36,880
changing a promise date, or securing outside capacity.

126
00:04:36,880 --> 00:04:39,840
Near term scheduling works with more facts and less room to move.

127
00:04:39,840 --> 00:04:42,360
The work orders exist, material status is clearer

128
00:04:42,360 --> 00:04:45,080
and machine conditions and staffing have a direct effect.

129
00:04:45,080 --> 00:04:47,320
At that point, the plan has to help people decide

130
00:04:47,320 --> 00:04:49,040
what happens on the floor today and tomorrow.

131
00:04:49,040 --> 00:04:51,160
That's why production planning isn't a calendar.

132
00:04:51,160 --> 00:04:53,920
It's the process of turning customer demand into work

133
00:04:53,920 --> 00:04:55,760
that the factory can physically complete

134
00:04:55,760 --> 00:04:58,520
while dealing with limited capacity, changing conditions,

135
00:04:58,520 --> 00:05:00,040
and competing priorities.

136
00:05:00,040 --> 00:05:02,280
Production planning software becomes interesting

137
00:05:02,280 --> 00:05:04,320
when that process involves more constraints

138
00:05:04,320 --> 00:05:06,880
and more changes than people can reliably hold together

139
00:05:06,880 --> 00:05:10,640
through calls, memory, whiteboards, and spreadsheets alone.

140
00:05:10,640 --> 00:05:12,440
Still, that doesn't mean every factory needs it.

141
00:05:12,440 --> 00:05:16,280
The question is whether your factory has reached that point yet.

142
00:05:16,280 --> 00:05:18,400
When spreadsheets and experience still work,

143
00:05:18,400 --> 00:05:21,120
here's something most planning software vendors don't mention,

144
00:05:21,120 --> 00:05:24,160
a dedicated planning system isn't always the right next step.

145
00:05:24,160 --> 00:05:26,160
In some plants, a spreadsheet, a planning board,

146
00:05:26,160 --> 00:05:28,680
and people who know the work handle the job just fine.

147
00:05:28,680 --> 00:05:30,760
Take a factory with a stable product range,

148
00:05:30,760 --> 00:05:33,600
repeatable routes, and demand that shift slowly.

149
00:05:33,600 --> 00:05:35,880
The same products run through the same work centers,

150
00:05:35,880 --> 00:05:37,520
lead times stay predictable,

151
00:05:37,520 --> 00:05:41,080
and there aren't many decisions about where each operation can land.

152
00:05:41,080 --> 00:05:43,280
In that setting, the planner sees the whole picture

153
00:05:43,280 --> 00:05:46,160
without chasing facts across five systems all day.

154
00:05:46,160 --> 00:05:48,000
That kind of manual planning has a strength

155
00:05:48,000 --> 00:05:49,920
most software projects overlook.

156
00:05:49,920 --> 00:05:51,280
It stays close to the work.

157
00:05:51,280 --> 00:05:54,120
A planner can walk to production, talk to a supervisor,

158
00:05:54,120 --> 00:05:55,680
and adjust the day around something

159
00:05:55,680 --> 00:05:57,920
the formal data hasn't caught up with yet.

160
00:05:57,920 --> 00:06:00,400
A paper board often beats a complex application

161
00:06:00,400 --> 00:06:02,000
when the team only needs to decide

162
00:06:02,000 --> 00:06:04,440
which of a handful of jobs should run after lunch.

163
00:06:04,440 --> 00:06:06,640
Excel also works fine for simple load checks,

164
00:06:06,640 --> 00:06:09,280
material follow-up, and short term coordination,

165
00:06:09,280 --> 00:06:12,200
especially when the same people plan and supervise the work,

166
00:06:12,200 --> 00:06:13,680
there's nothing outdated about that.

167
00:06:13,680 --> 00:06:15,120
Factories don't need more technology

168
00:06:15,120 --> 00:06:18,120
just because a software vendor can draw a tidy gant chart.

169
00:06:18,120 --> 00:06:19,360
You need a better planning method

170
00:06:19,360 --> 00:06:21,680
only when the current one stops producing decisions

171
00:06:21,680 --> 00:06:22,520
people can trust.

172
00:06:22,520 --> 00:06:23,760
A whiteboard isn't the problem,

173
00:06:23,760 --> 00:06:25,400
and a spreadsheet isn't the problem.

174
00:06:25,400 --> 00:06:26,880
The real question is whether the method

175
00:06:26,880 --> 00:06:30,000
still matches the amount of variation the plant has to absorb.

176
00:06:30,000 --> 00:06:33,040
Think about a small machining area with a limited set of repeat customers.

177
00:06:33,040 --> 00:06:34,760
The team knows the parts, knows the machines,

178
00:06:34,760 --> 00:06:37,120
and knows which jobs normally run well together.

179
00:06:37,120 --> 00:06:39,360
One planner works closely with the production lead,

180
00:06:39,360 --> 00:06:41,080
and each morning they review the queue,

181
00:06:41,080 --> 00:06:43,760
check for anything unusual, then set the order for the day.

182
00:06:43,760 --> 00:06:45,320
That's a sensible control loop.

183
00:06:45,320 --> 00:06:49,520
It works because the planning knowledge stays close to the people doing the work,

184
00:06:49,520 --> 00:06:52,040
and because the number of changes stays low enough

185
00:06:52,040 --> 00:06:54,240
that they can think through the consequences,

186
00:06:54,240 --> 00:06:56,560
they don't need a mathematical optimizer to tell them

187
00:06:56,560 --> 00:06:58,400
that two jobs should run back to back,

188
00:06:58,400 --> 00:07:01,000
because they use the same fixture, they already know it,

189
00:07:01,000 --> 00:07:04,200
and the decision only affects a small set of work orders.

190
00:07:04,200 --> 00:07:07,840
Manual planning also works where the production flow itself creates stability.

191
00:07:07,840 --> 00:07:10,880
A repetitive line with a fixed sequence, stable cycle times,

192
00:07:10,880 --> 00:07:13,800
and a limited number of variants doesn't need the same detailed scheduling

193
00:07:13,800 --> 00:07:15,480
as a high mix job shop.

194
00:07:15,480 --> 00:07:18,200
The line already contains most of the plan.

195
00:07:18,200 --> 00:07:19,920
Material moves in a known direction,

196
00:07:19,920 --> 00:07:23,040
capacity follows a known rhythm and deviations show up fast.

197
00:07:23,040 --> 00:07:27,360
You might still need ERP, MES, quality data, and maintenance planning,

198
00:07:27,360 --> 00:07:31,040
but you don't necessarily need another layer to choose every next operation.

199
00:07:31,040 --> 00:07:33,120
The same goes for some make-to-order businesses,

200
00:07:33,120 --> 00:07:35,480
even when the work looks complex from the outside.

201
00:07:35,480 --> 00:07:38,160
If one experienced planner manages a modest order volume,

202
00:07:38,160 --> 00:07:39,960
the roots stay mostly consistent,

203
00:07:39,960 --> 00:07:42,040
and the plant has enough slack capacity,

204
00:07:42,040 --> 00:07:45,800
then direct coordination can stay faster than system-driven planning.

205
00:07:45,800 --> 00:07:47,600
That planner isn't relying on luck,

206
00:07:47,600 --> 00:07:50,560
they're applying judgment in an environment where judgment can still keep up.

207
00:07:50,560 --> 00:07:52,680
Now, here's what manual planning really depends on.

208
00:07:52,680 --> 00:07:54,800
It depends on direct access to information,

209
00:07:54,800 --> 00:07:57,000
on people knowing each other's constraints,

210
00:07:57,000 --> 00:08:00,880
and on changes arriving slowly enough that the plan survives long enough to be useful.

211
00:08:00,880 --> 00:08:03,440
And it depends on the planner having time to think,

212
00:08:03,440 --> 00:08:06,960
rather than spending every hour answering calls and correcting dates.

213
00:08:06,960 --> 00:08:10,280
When those conditions hold, simple tools work extremely well.

214
00:08:10,280 --> 00:08:13,560
The problem starts when a spreadsheet becomes a shared operational database,

215
00:08:13,560 --> 00:08:15,840
a scheduling engine, an approval workflow,

216
00:08:15,840 --> 00:08:18,160
and the only place anyone can find the current truth.

217
00:08:18,160 --> 00:08:22,320
That's when you start seeing version names like Final Plan V12 really final.

218
00:08:22,320 --> 00:08:25,200
Manufacturing has survived worse, but that's not a control method.

219
00:08:25,200 --> 00:08:28,880
A manual plan gets fragile when its speed comes from one person remembering rules

220
00:08:28,880 --> 00:08:33,040
that nobody else can inspect, or when the board in one area doesn't match the priorities in another.

221
00:08:33,040 --> 00:08:37,040
It gets even more fragile when a small disruption forces someone to call three people,

222
00:08:37,040 --> 00:08:40,800
open four files, and guess which promise date still means anything.

223
00:08:40,800 --> 00:08:43,160
That boundary doesn't arrive at a fixed company size.

224
00:08:43,160 --> 00:08:46,320
You can run a large but stable plan with fairly simple planning methods,

225
00:08:46,320 --> 00:08:49,200
and you can also run a small plant where shared machines,

226
00:08:49,200 --> 00:08:52,720
changing routes, short customer lead times, and constant exceptions,

227
00:08:52,720 --> 00:08:55,280
turn every day into a planning puzzle.

228
00:08:55,280 --> 00:08:57,280
The issue isn't whether people use Excel.

229
00:08:57,280 --> 00:09:00,320
It's whether Excel and experience can still keep pace with change.

230
00:09:00,320 --> 00:09:04,000
When changes start arriving faster than the planning team can test the consequences,

231
00:09:04,000 --> 00:09:08,400
the same tools that once gave people control start creating delay, uncertainty,

232
00:09:08,400 --> 00:09:10,160
and informal workarounds.

233
00:09:10,160 --> 00:09:13,920
The first warning sign, plans change faster than people can replan.

234
00:09:13,920 --> 00:09:15,360
Here's the first warning sign.

235
00:09:15,360 --> 00:09:18,320
The plan changes so often that nobody has time to plan anymore.

236
00:09:18,320 --> 00:09:21,520
A rush order from sales, a customer pulling a delivery date forward,

237
00:09:21,520 --> 00:09:24,160
a supplier missing a component, someone calling in sick,

238
00:09:24,160 --> 00:09:26,320
a machine stopping halfway through a job.

239
00:09:26,320 --> 00:09:28,320
None of those events are unusual on their own.

240
00:09:28,320 --> 00:09:30,640
Manufacturing has always dealt with disruption.

241
00:09:30,640 --> 00:09:33,920
The trouble starts when several disruptions hit during the same week

242
00:09:33,920 --> 00:09:36,400
and each one forces a new chain of decisions.

243
00:09:36,400 --> 00:09:38,000
Changing a date takes seconds,

244
00:09:38,000 --> 00:09:41,840
but testing whether that new date creates a workable plan takes much longer.

245
00:09:41,840 --> 00:09:44,240
The plan has to trace the effect through operations,

246
00:09:44,240 --> 00:09:46,720
resource availability, material status,

247
00:09:46,720 --> 00:09:49,440
shift patterns, and other customer commitments.

248
00:09:49,440 --> 00:09:52,320
If the only response is to move the order higher in a spreadsheet,

249
00:09:52,320 --> 00:09:54,000
you haven't solved the planning problem.

250
00:09:54,000 --> 00:09:55,920
You've just pushed the question downstream.

251
00:09:55,920 --> 00:09:59,200
Usually to the supervisor who now has to explain why two urgent orders

252
00:09:59,200 --> 00:10:00,800
both need the same machine.

253
00:10:00,800 --> 00:10:03,760
Let's walk through a scenario from a constrained production week

254
00:10:03,760 --> 00:10:05,520
in a hypothetical machining plant,

255
00:10:05,520 --> 00:10:08,480
the plant committed several orders for shipment by the end of the week.

256
00:10:08,480 --> 00:10:11,760
On Tuesday morning, a machining center stops with a fault.

257
00:10:11,760 --> 00:10:14,320
Maintenance expects it to stay down for several hours,

258
00:10:14,320 --> 00:10:16,560
though nobody can promise the exact time yet.

259
00:10:16,560 --> 00:10:18,880
That machine doesn't process just one order.

260
00:10:18,880 --> 00:10:21,120
It sits on the route for several jobs,

261
00:10:21,120 --> 00:10:23,360
including a part that finally assembly needs.

262
00:10:23,360 --> 00:10:25,280
The plan now faces real decisions,

263
00:10:25,280 --> 00:10:27,520
can the affected work move to another machine?

264
00:10:27,520 --> 00:10:29,440
If so, does that machine need a different fixture,

265
00:10:29,440 --> 00:10:30,800
program, or operator?

266
00:10:30,800 --> 00:10:33,280
Does moving it bump another order with a customer commitment?

267
00:10:33,280 --> 00:10:34,480
Can the team split the batch?

268
00:10:34,480 --> 00:10:36,480
Is their material ready for an alternate route?

269
00:10:36,480 --> 00:10:38,880
And if the breakdown lasts longer than expected,

270
00:10:38,880 --> 00:10:41,520
when does the plant need to tell sales that one promise date

271
00:10:41,520 --> 00:10:42,720
is no longer realistic?

272
00:10:42,720 --> 00:10:44,400
Those questions arrive before lunch.

273
00:10:44,400 --> 00:10:47,920
In a simple environment, an experienced planner can work through them.

274
00:10:47,920 --> 00:10:50,240
They talk to maintenance, production, quality,

275
00:10:50,240 --> 00:10:52,640
and purchasing, then decide what to do.

276
00:10:52,640 --> 00:10:55,440
But as order volume rises, the plant runs more shifts,

277
00:10:55,440 --> 00:10:57,040
and shared resources multiply,

278
00:10:57,040 --> 00:10:59,680
that same planner spends less time building a solid plan,

279
00:10:59,680 --> 00:11:01,360
and more time patching up the last one.

280
00:11:01,360 --> 00:11:02,720
That's a shift that matters.

281
00:11:02,720 --> 00:11:04,400
Planning becomes damage control.

282
00:11:04,400 --> 00:11:07,120
The official schedule turns into a starting suggestion,

283
00:11:07,120 --> 00:11:09,280
while the shop floor follows calls, messages,

284
00:11:09,280 --> 00:11:10,320
and local judgment,

285
00:11:10,320 --> 00:11:12,480
because people know the printed plan

286
00:11:12,480 --> 00:11:14,320
no longer matches reality.

287
00:11:14,320 --> 00:11:16,320
You might still publish a schedule each morning,

288
00:11:16,320 --> 00:11:18,720
but by mid-shift, it has already lost authority.

289
00:11:18,720 --> 00:11:21,840
When a plan loses authority, informal priorities take over.

290
00:11:21,840 --> 00:11:23,600
A production supervisor may run the work

291
00:11:23,600 --> 00:11:24,960
that keeps the machine busy,

292
00:11:24,960 --> 00:11:27,600
a sales manager may push the loudest customer request,

293
00:11:27,600 --> 00:11:29,680
and a material coordinator may hold work

294
00:11:29,680 --> 00:11:31,920
because one component looks uncertain.

295
00:11:31,920 --> 00:11:35,200
Each decision sounds reasonable from where that person sits.

296
00:11:35,200 --> 00:11:38,080
The problem appears when nobody can see the combined effect.

297
00:11:38,080 --> 00:11:39,760
This isn't a knock on the people involved,

298
00:11:39,760 --> 00:11:41,600
they're responding to what they know.

299
00:11:41,600 --> 00:11:44,400
But the factory now runs on a stream of local corrections,

300
00:11:44,400 --> 00:11:46,400
rather than a shared, tested plan.

301
00:11:46,400 --> 00:11:48,240
Planets become human message brokers.

302
00:11:48,240 --> 00:11:50,800
They ask for updates, compare dates, make calls,

303
00:11:50,800 --> 00:11:53,520
amend spreadsheets, send revised priorities,

304
00:11:53,520 --> 00:11:55,520
and then start over after the next disruption.

305
00:11:55,520 --> 00:11:57,440
The work feels busy because it is busy,

306
00:11:57,440 --> 00:12:00,080
yet the plant may still struggle to answer a basic question.

307
00:12:00,080 --> 00:12:01,520
What should run next and why?

308
00:12:01,520 --> 00:12:03,280
Let me separate two types of replanning.

309
00:12:03,280 --> 00:12:05,440
The first type is administrative replanning.

310
00:12:05,440 --> 00:12:08,240
Someone changes a due date, moves an order to a new day,

311
00:12:08,240 --> 00:12:10,080
or adds a note that the job is urgent.

312
00:12:10,080 --> 00:12:11,360
That keeps records current,

313
00:12:11,360 --> 00:12:14,320
but it doesn't prove the revised plan can actually happen.

314
00:12:14,320 --> 00:12:16,720
The second type is feasibility-based replanning.

315
00:12:16,720 --> 00:12:20,080
The plan attests a change against the actual rules governing production,

316
00:12:20,080 --> 00:12:21,920
then compares possible outcomes.

317
00:12:21,920 --> 00:12:24,000
One option might protect a high priority shipment,

318
00:12:24,000 --> 00:12:26,080
but delay two lower priority orders.

319
00:12:26,080 --> 00:12:28,800
Another might preserve setup flow, but require overtime.

320
00:12:28,800 --> 00:12:30,560
A third might wait for a material delivery

321
00:12:30,560 --> 00:12:32,160
because producing the earlier operation

322
00:12:32,160 --> 00:12:33,920
now doesn't improve the final ship date.

323
00:12:33,920 --> 00:12:35,920
That is the work people need support with.

324
00:12:35,920 --> 00:12:38,720
Production planning software doesn't remove disruption.

325
00:12:38,720 --> 00:12:39,680
Nothing does.

326
00:12:39,680 --> 00:12:42,000
But it gives the planner a way to test the consequences

327
00:12:42,000 --> 00:12:43,360
of disruption faster,

328
00:12:43,360 --> 00:12:45,760
using rules the team can inspect and challenge.

329
00:12:45,760 --> 00:12:47,440
When the system can show which orders,

330
00:12:47,440 --> 00:12:49,600
resources, and dates, a change affects,

331
00:12:49,600 --> 00:12:52,400
the conversation moves from guesswork to trade-offs.

332
00:12:52,400 --> 00:12:56,720
Still, faster replanning only helps if the plan understands what capacity really means,

333
00:12:56,720 --> 00:12:59,680
because many factories appear to have free capacity in the system

334
00:12:59,680 --> 00:13:01,760
right up until they try to use it.

335
00:13:01,760 --> 00:13:03,280
The second warning sign.

336
00:13:03,280 --> 00:13:06,160
Capacity exists on paper, but not in practice.

337
00:13:06,160 --> 00:13:08,320
Here's the next warning sign, and it's a tricky one,

338
00:13:08,320 --> 00:13:09,760
the kind that creeps up on you.

339
00:13:09,760 --> 00:13:11,600
Your system says there's available capacity,

340
00:13:11,600 --> 00:13:13,600
but the shop floor keeps burning overtime,

341
00:13:13,600 --> 00:13:16,880
expediting everything or shuffling work at the last minute just to hit the plan.

342
00:13:17,760 --> 00:13:20,880
On paper, a machine might show 43 hours next week.

343
00:13:20,880 --> 00:13:23,280
That number treats every hour as equally usable,

344
00:13:23,280 --> 00:13:24,720
but production knows better.

345
00:13:24,720 --> 00:13:26,240
Some of those hours go to setups,

346
00:13:26,240 --> 00:13:27,760
others belong to planned maintenance.

347
00:13:27,760 --> 00:13:29,600
The machine might need a specific tool,

348
00:13:29,600 --> 00:13:32,960
a program approval or an operator who only works one shift.

349
00:13:32,960 --> 00:13:34,400
Material might arrive at the building,

350
00:13:34,400 --> 00:13:36,240
but still sit waiting for inspection.

351
00:13:36,240 --> 00:13:38,160
So the capacity number looks comforting,

352
00:13:38,160 --> 00:13:39,280
the plan doesn't match it.

353
00:13:39,280 --> 00:13:42,240
Take two CNC machines that can both machine the same part.

354
00:13:42,240 --> 00:13:44,560
ERP might group them under one work center,

355
00:13:44,560 --> 00:13:46,400
which works fine for rough planning.

356
00:13:46,400 --> 00:13:48,240
The plan is sees open hours on machine B

357
00:13:48,240 --> 00:13:50,800
and assumes work can shift when machine A gets busy,

358
00:13:50,800 --> 00:13:54,480
but machine B might not have the probing setup that customer requires.

359
00:13:54,480 --> 00:13:56,320
Or its current fixture can't hold the part,

360
00:13:56,320 --> 00:13:57,280
it might run the part,

361
00:13:57,280 --> 00:14:00,080
but only with a qualified operator who won't be free until Thursday.

362
00:14:00,080 --> 00:14:01,360
The program could exist,

363
00:14:01,360 --> 00:14:04,800
but engineering hasn't released the latest revision to that machine yet.

364
00:14:04,800 --> 00:14:06,320
That's not spare capacity.

365
00:14:06,320 --> 00:14:08,720
It's a theoretical possibility with a list of conditions

366
00:14:08,720 --> 00:14:10,160
that nobody wrote down.

367
00:14:10,160 --> 00:14:12,800
Now, here's where finite capacity planning comes in.

368
00:14:12,800 --> 00:14:15,040
The phrase sounds more complicated than it is.

369
00:14:15,040 --> 00:14:17,520
Finite capacity planning just means the schedule accepts

370
00:14:17,520 --> 00:14:20,640
that a resource can only do so much work in a real period of time.

371
00:14:20,640 --> 00:14:23,040
If a machine already has work booked from eight in the morning

372
00:14:23,040 --> 00:14:24,560
until four in the afternoon,

373
00:14:24,560 --> 00:14:25,920
the next job can't start at nine,

374
00:14:25,920 --> 00:14:28,880
just because an ERP lead time calculation wants it to.

375
00:14:28,880 --> 00:14:30,720
The same thinking applies beyond machines.

376
00:14:30,720 --> 00:14:31,920
A resource could be a furnace,

377
00:14:31,920 --> 00:14:33,200
a test station, a specialist,

378
00:14:33,200 --> 00:14:35,840
a shared tool, a packing cell or an inspection step.

379
00:14:35,840 --> 00:14:37,920
Any of those can restrict the flow of work,

380
00:14:37,920 --> 00:14:40,320
and the plan needs to treat them as real limits,

381
00:14:40,320 --> 00:14:42,800
not as vague assumptions for production to figure out.

382
00:14:42,800 --> 00:14:46,720
Now, a finite capacity model doesn't need to capture every second of every shift.

383
00:14:46,720 --> 00:14:49,200
That approach usually turns into a master data project

384
00:14:49,200 --> 00:14:50,560
with a planning tool attached,

385
00:14:50,560 --> 00:14:52,320
and nobody enjoys that outcome.

386
00:14:52,320 --> 00:14:53,840
It needs to capture the constraints

387
00:14:53,840 --> 00:14:55,760
that regularly change your decisions.

388
00:14:55,760 --> 00:14:56,960
Maybe set up time matters

389
00:14:56,960 --> 00:14:59,200
because product families need to run together.

390
00:14:59,200 --> 00:15:02,000
Maybe a heat treatment batch needs a minimum and maximum load.

391
00:15:02,000 --> 00:15:04,400
Maybe only one technician can approve a certain test

392
00:15:04,400 --> 00:15:07,200
or maintenance blocks a machine every Wednesday afternoon.

393
00:15:07,200 --> 00:15:10,640
If these conditions repeatedly create late orders or urgent changes,

394
00:15:10,640 --> 00:15:12,240
they belong in the planning logic.

395
00:15:12,240 --> 00:15:14,000
Consider a plant with a forming process

396
00:15:14,000 --> 00:15:16,000
and a downstream assembly area.

397
00:15:16,000 --> 00:15:18,800
The forming machine runs with strong OEE,

398
00:15:18,800 --> 00:15:20,800
overall equipment effectiveness.

399
00:15:20,800 --> 00:15:22,800
And the team takes pride in keeping it busy.

400
00:15:22,800 --> 00:15:24,480
It produces parts all day,

401
00:15:24,480 --> 00:15:27,680
but downstream assembly can only handle a limited number of product variants

402
00:15:27,680 --> 00:15:28,960
before a changeover,

403
00:15:28,960 --> 00:15:30,800
and one component for the next assembly batch

404
00:15:30,800 --> 00:15:32,480
hasn't cleared incoming inspection.

405
00:15:32,480 --> 00:15:35,040
The forming machine keeps producing because it has capacity.

406
00:15:35,040 --> 00:15:36,640
The factory builds work in progress

407
00:15:36,640 --> 00:15:39,040
because the next step can't use those parts yet.

408
00:15:39,040 --> 00:15:40,560
Local capacity looks good,

409
00:15:40,560 --> 00:15:42,320
system capacity doesn't.

410
00:15:42,320 --> 00:15:44,880
This is why broad capacity buckets can mislead you.

411
00:15:44,880 --> 00:15:47,200
They show available hours by work centre for each week,

412
00:15:47,200 --> 00:15:49,760
but they don't show the sequence of work the tools required,

413
00:15:49,760 --> 00:15:51,920
the time lost when switching product families,

414
00:15:51,920 --> 00:15:54,640
or the point where work waits for a downstream operation.

415
00:15:54,640 --> 00:15:57,280
A bucket can tell you that capacity exists somewhere,

416
00:15:57,280 --> 00:15:59,760
but it can't always tell you whether that capacity helps

417
00:15:59,760 --> 00:16:01,440
the order currently at risk.

418
00:16:01,440 --> 00:16:04,960
You often spot this issue through behaviour before you see it in a report.

419
00:16:04,960 --> 00:16:06,400
The plant keeps adding overtime

420
00:16:06,400 --> 00:16:08,400
even though the planning view shows room.

421
00:16:08,400 --> 00:16:10,720
Supervisors ask for work to be released early,

422
00:16:10,720 --> 00:16:12,880
because they don't trust later capacity.

423
00:16:12,880 --> 00:16:16,320
Purchasing expedites material for orders that production can't start yet.

424
00:16:16,320 --> 00:16:18,480
The planar spends time hunting for a machine slot

425
00:16:18,480 --> 00:16:22,640
that supposedly exists only to find it comes with conditions nobody captured.

426
00:16:22,640 --> 00:16:26,240
That patent points to a gap between nominal capacity and usable capacity.

427
00:16:26,240 --> 00:16:29,520
In practical terms, nominal capacity is what the calendar says,

428
00:16:29,520 --> 00:16:33,280
while usable capacity is what the factory can actually turn into finished product

429
00:16:33,280 --> 00:16:34,640
under its current rules.

430
00:16:34,640 --> 00:16:37,040
People, material status and resource limits.

431
00:16:37,680 --> 00:16:39,920
Production planning software can help with that gap,

432
00:16:39,920 --> 00:16:43,360
but only if the model represents the limits people work around every day.

433
00:16:43,360 --> 00:16:46,000
It can test whether an operation fits on an approved resource,

434
00:16:46,000 --> 00:16:47,360
respect a resource calendar,

435
00:16:47,360 --> 00:16:48,800
consider a setup sequence,

436
00:16:48,800 --> 00:16:51,760
and expose the orders displaced when you assign a job elsewhere.

437
00:16:51,760 --> 00:16:53,600
It can't invent a qualified operator.

438
00:16:53,600 --> 00:16:55,360
It can't create a missing fixture,

439
00:16:55,360 --> 00:16:57,600
and it shouldn't pretend an open time slot means much

440
00:16:57,600 --> 00:16:59,920
when a material shortage blocks the next operation.

441
00:16:59,920 --> 00:17:03,280
Once the plan starts treating capacity as a set of real production constraints,

442
00:17:03,280 --> 00:17:06,720
people can stop arguing about whether hours exist somewhere in the system.

443
00:17:06,720 --> 00:17:08,400
They can ask the more useful question,

444
00:17:08,400 --> 00:17:10,640
which constraint stops this order from moving,

445
00:17:10,640 --> 00:17:13,200
and what choice creates the least damage elsewhere.

446
00:17:13,200 --> 00:17:14,400
The third warning sign,

447
00:17:14,400 --> 00:17:17,120
the same order has different dates in different systems.

448
00:17:17,120 --> 00:17:19,600
Here's another warning sign, and this one feels personal.

449
00:17:19,600 --> 00:17:21,680
The same order seems to have several delivery dates,

450
00:17:21,680 --> 00:17:22,960
depending on who you ask.

451
00:17:22,960 --> 00:17:25,120
Sales sees the customer promise date in ERP.

452
00:17:25,120 --> 00:17:27,520
The planner has a revised date in a spreadsheet.

453
00:17:27,520 --> 00:17:30,080
The manufacturing execution system, or MS,

454
00:17:30,080 --> 00:17:32,240
shows that part of the order finished one operation.

455
00:17:32,240 --> 00:17:35,840
Production has a whiteboard date based on what the supervisor things can run next.

456
00:17:35,840 --> 00:17:39,120
Purchasing works from a supplier date that changed the yesterday afternoon.

457
00:17:39,120 --> 00:17:40,960
Every date probably has a reason behind it.

458
00:17:40,960 --> 00:17:44,080
The trouble starts when nobody can explain which date

459
00:17:44,080 --> 00:17:45,840
drives the next production decision,

460
00:17:45,840 --> 00:17:47,840
and what assumptions sit behind that date.

461
00:17:47,840 --> 00:17:51,280
Then the planning meeting stops being about how to protect customer delivery.

462
00:17:51,280 --> 00:17:53,120
It becomes a meeting where people compare records

463
00:17:53,120 --> 00:17:56,080
and try to work out which version of the order is closest to the truth.

464
00:17:56,080 --> 00:17:58,880
That waste's time and worse, it removes trust.

465
00:17:58,880 --> 00:18:00,000
Let's make it concrete.

466
00:18:00,000 --> 00:18:02,240
Imagine an order for 100 assemblies.

467
00:18:02,240 --> 00:18:05,360
The MS records that 60 components completed machining,

468
00:18:05,360 --> 00:18:08,400
ERP still shows the full production order as open,

469
00:18:08,400 --> 00:18:10,720
because the finished assemblies haven't reached stock.

470
00:18:10,720 --> 00:18:14,880
The planner's file treats the 60 completed components as available for the next operation,

471
00:18:14,880 --> 00:18:17,360
because that's how the plan looked at the start of the day.

472
00:18:17,360 --> 00:18:20,000
But quality has held 10 of those components for review.

473
00:18:20,000 --> 00:18:22,880
Now each system reports something technically defensible,

474
00:18:22,880 --> 00:18:25,280
yet they don't describe the same planning situation.

475
00:18:25,280 --> 00:18:28,880
The MS reports execution, ERP reports the order at its business stage.

476
00:18:28,880 --> 00:18:31,840
The spreadsheet reports an assumption about what work can move forward.

477
00:18:31,840 --> 00:18:34,240
None of that becomes useful until the planner can connect

478
00:18:34,240 --> 00:18:37,680
quantity, quality status, routing step, material state

479
00:18:37,680 --> 00:18:40,160
and the remaining capacity needed to finish the order.

480
00:18:40,160 --> 00:18:43,440
A date conflict doesn't always mean bad data, factories change.

481
00:18:43,440 --> 00:18:45,920
An order may move because a customer changes a request,

482
00:18:45,920 --> 00:18:48,400
a machine needs repair, or a supplier slips.

483
00:18:48,400 --> 00:18:50,640
The systems won't all update at the same second,

484
00:18:50,640 --> 00:18:52,320
and they don't need to.

485
00:18:52,320 --> 00:18:53,840
The problem isn't that change exists.

486
00:18:53,840 --> 00:18:56,000
The problem is when changes have no shared process,

487
00:18:56,000 --> 00:19:00,000
no clear owner, and no way to test whether the revised date remains physically possible.

488
00:19:00,000 --> 00:19:02,640
Think about the questions people ask in this situation.

489
00:19:02,640 --> 00:19:04,560
Can we still ship this on Thursday?

490
00:19:04,560 --> 00:19:06,080
What's the real finished date?

491
00:19:06,080 --> 00:19:09,840
Why does the dashboard say green when production says the order is at risk?

492
00:19:09,840 --> 00:19:13,840
Those aren't reporting questions, they are planning questions disguised as date questions.

493
00:19:13,840 --> 00:19:16,960
A dashboard can show several dates side by side.

494
00:19:16,960 --> 00:19:19,760
That may help expose the problem, but it doesn't settle it.

495
00:19:19,760 --> 00:19:24,160
The team still needs a planning rule that answers which date comes from a feasible sequence

496
00:19:24,160 --> 00:19:26,640
of remaining work under the current constraints.

497
00:19:26,640 --> 00:19:30,720
Without that, an ERP due date can look firm even when it came from a fixed lead time.

498
00:19:30,720 --> 00:19:34,880
A plan is date can look practical even when it assumes material that quality hasn't released.

499
00:19:34,880 --> 00:19:38,880
And a sales promise can travel through the business without anyone testing it against the shop floor.

500
00:19:38,880 --> 00:19:42,240
That's how date drift becomes normal, people start using qualifiers.

501
00:19:42,240 --> 00:19:43,520
That's the system date.

502
00:19:43,520 --> 00:19:44,960
That's the planning date.

503
00:19:44,960 --> 00:19:46,960
That's the date we told the customer.

504
00:19:46,960 --> 00:19:51,760
Each qualifier signals that the factory lacks one shared view of what the order can actually do next.

505
00:19:51,760 --> 00:19:54,400
You don't solve this by forcing every system to own every date.

506
00:19:54,400 --> 00:19:57,200
ERP should still own commercial commitments and order data.

507
00:19:57,200 --> 00:19:59,520
MS should still record what happened on the floor.

508
00:19:59,520 --> 00:20:04,080
Planning needs its own logic for evaluating remaining work, resource limits and likely completion.

509
00:20:04,080 --> 00:20:08,160
The answer is a connected flow where execution facts update the planning view,

510
00:20:08,160 --> 00:20:12,240
planning consequences inform the business, and each date keeps a clear meaning.

511
00:20:12,240 --> 00:20:15,440
In practical terms, that means an order needs more than a due date.

512
00:20:15,440 --> 00:20:16,640
It needs context.

513
00:20:16,640 --> 00:20:18,160
Which operations remain?

514
00:20:18,160 --> 00:20:19,520
What quantity past quality?

515
00:20:19,520 --> 00:20:20,640
Which material is ready?

516
00:20:20,640 --> 00:20:22,640
Which resource can perform the next step?

517
00:20:22,640 --> 00:20:24,400
What work already claims that resource?

518
00:20:24,400 --> 00:20:25,600
And if the plan changes?

519
00:20:25,600 --> 00:20:26,960
Which promised dates?

520
00:20:26,960 --> 00:20:27,600
Move with it.

521
00:20:28,320 --> 00:20:30,080
Once people can answer those questions,

522
00:20:30,080 --> 00:20:31,840
dates stop competing with each other.

523
00:20:31,840 --> 00:20:34,560
They become different views of the same operational situation.

524
00:20:34,560 --> 00:20:36,080
That brings us to the deeper issue.

525
00:20:36,080 --> 00:20:39,600
When dates keep drifting, the factory usually doesn't just lack a better report.

526
00:20:39,600 --> 00:20:44,080
It lacks the planning context that connects an order to the work, resources and conditions

527
00:20:44,080 --> 00:20:46,000
that determine whether the date can hold.

528
00:20:46,000 --> 00:20:49,200
The fourth warning sign, bottle necks move but the plan doesn't.

529
00:20:49,200 --> 00:20:50,720
Here's warning sign number four.

530
00:20:50,720 --> 00:20:53,520
You're treating one resource as the bottle neck.

531
00:20:53,520 --> 00:20:57,680
But the real constraint has already moved somewhere else, and you haven't noticed yet.

532
00:20:57,680 --> 00:20:59,440
Most plants know their usual bottle neck.

533
00:20:59,440 --> 00:21:03,440
A paint line, a heat treatment furnace, a machining center, a test cell.

534
00:21:03,440 --> 00:21:06,720
They plan around it because it often limits output and that knowledge helps.

535
00:21:06,720 --> 00:21:11,280
But it can become a fixed belief, and fixed beliefs don't handle changing production conditions very well.

536
00:21:11,280 --> 00:21:15,120
The bottle neck you learned six months ago might not be the bottle neck you're fighting right now.

537
00:21:15,120 --> 00:21:18,240
A bottle neck is just the point that limits flow at this moment.

538
00:21:18,240 --> 00:21:20,320
And that point can change with product mix,

539
00:21:20,320 --> 00:21:24,640
machine condition, rework, a missing component, or who's on shift today.

540
00:21:24,640 --> 00:21:28,640
A resource that usually has spare time can become the constraint when several orders suddenly need

541
00:21:28,640 --> 00:21:32,640
the same inspection, the same tool, the same skill, or the same downstream assembly step.

542
00:21:32,640 --> 00:21:34,320
It happens fast.

543
00:21:34,320 --> 00:21:36,320
Take a plant that makes several configured products.

544
00:21:36,320 --> 00:21:38,640
They all move through machining then into final assembly.

545
00:21:38,640 --> 00:21:42,080
Normally machining gets most of the attention because it runs close to full load,

546
00:21:42,080 --> 00:21:44,880
so the team works hard to keep those machines producing.

547
00:21:44,880 --> 00:21:46,800
But one week the order mix shifts.

548
00:21:46,800 --> 00:21:49,600
Suddenly more products need a special inspection after machining.

549
00:21:49,600 --> 00:21:52,080
Only one inspection station has the right test setup.

550
00:21:52,080 --> 00:21:54,320
And at the same time a few parts need rework.

551
00:21:54,320 --> 00:21:56,320
Machining keeps producing at a strong pace,

552
00:21:56,320 --> 00:22:00,640
but finished parts start queuing before inspection and final assembly has nothing to work with.

553
00:22:00,640 --> 00:22:04,080
Even though machining looks busy and productive.

554
00:22:04,080 --> 00:22:05,280
The constraint has moved.

555
00:22:05,280 --> 00:22:07,920
If your plant still treats machining as the only bottle neck,

556
00:22:07,920 --> 00:22:10,000
it'll keep releasing more work into the same route.

557
00:22:10,000 --> 00:22:13,280
That builds queues, fills floor space, ties up material in work in progress.

558
00:22:13,280 --> 00:22:14,960
Nobody set out to create that mess,

559
00:22:14,960 --> 00:22:17,440
but the plant simply followed an old assumption

560
00:22:17,440 --> 00:22:19,200
after the operating conditions changed.

561
00:22:19,200 --> 00:22:23,280
Now overall equipment effectiveness, OEE, often gets misunderstood here.

562
00:22:23,280 --> 00:22:26,240
OEE tells you useful things about how a machine performs.

563
00:22:26,240 --> 00:22:28,720
It looks at availability, performance, and quality,

564
00:22:28,720 --> 00:22:31,120
and shows whether a machine loses time through stops,

565
00:22:31,120 --> 00:22:32,800
slow running, or rejected output.

566
00:22:32,800 --> 00:22:36,560
Production teams need that information because it helps them improve how a resource runs.

567
00:22:36,560 --> 00:22:39,200
But here's the thing, OEE doesn't produce a production plan.

568
00:22:39,200 --> 00:22:42,320
A machine can show strong OEE while the factory misses delivery dates,

569
00:22:42,320 --> 00:22:43,760
and that's not a contradiction.

570
00:22:43,760 --> 00:22:45,920
The machine may produce parts efficiently,

571
00:22:45,920 --> 00:22:48,480
but those parts could be waiting for a shared tool,

572
00:22:48,480 --> 00:22:51,760
a quality release, an assembly slot, or material for the next stage.

573
00:22:52,320 --> 00:22:54,560
Local machine performance and order flow are connected,

574
00:22:54,560 --> 00:22:56,480
but they answer different questions.

575
00:22:56,480 --> 00:23:00,640
Think about a high OEE cutting machine feeding a downstream coating process.

576
00:23:00,640 --> 00:23:03,760
The cutting machine stays loaded because its team wants to avoid idle time,

577
00:23:03,760 --> 00:23:07,040
so it produces a large volume of parts for several future orders.

578
00:23:07,040 --> 00:23:10,880
But the coating process has a limited number of color changes it can handle per shift,

579
00:23:10,880 --> 00:23:14,240
and an urgent customer order needs a color that isn't in the planned sequence.

580
00:23:14,240 --> 00:23:16,320
Now the plant faces a trade-off,

581
00:23:16,320 --> 00:23:18,880
it can keep the cutting machine busy with whatever comes next.

582
00:23:18,880 --> 00:23:21,760
That improves local output, but adds waiting stock,

583
00:23:21,760 --> 00:23:24,880
or it can change the upstream sequence to feed the coating process

584
00:23:24,880 --> 00:23:27,280
with the parts that protect customer delivery,

585
00:23:27,280 --> 00:23:30,880
even if that creates a less attractive setup or a short-gab at cutting.

586
00:23:30,880 --> 00:23:34,720
Neither option is perfect because production planning rarely gives you perfect options.

587
00:23:34,720 --> 00:23:36,560
It should give you visible consequences,

588
00:23:36,560 --> 00:23:39,600
so the plan needs to understand the root, not just the first resource.

589
00:23:39,600 --> 00:23:43,280
It needs to see which operations feed downstream constraints,

590
00:23:43,280 --> 00:23:45,440
which orders carry near-term commitments,

591
00:23:45,440 --> 00:23:48,640
and where releasing more work creates a queue rather than progress.

592
00:23:48,640 --> 00:23:50,800
And it needs regular feedback from execution,

593
00:23:50,800 --> 00:23:53,440
because yesterday's constraint may not limit flow today.

594
00:23:53,440 --> 00:23:57,920
That doesn't mean the planning system has to recalculate the entire plant every time a sensor changes state.

595
00:23:57,920 --> 00:24:00,640
That would create noise, and production already has enough noise.

596
00:24:00,640 --> 00:24:03,440
It means defined events that trigger a review,

597
00:24:03,440 --> 00:24:06,160
a major machine outage, a quality hold,

598
00:24:06,160 --> 00:24:10,080
a material delay, a staffing gap, or a shift in the order mix.

599
00:24:10,080 --> 00:24:12,720
The goal is to plan around the current constraint.

600
00:24:12,720 --> 00:24:15,280
Sometimes that constraint stays in the same place for months.

601
00:24:15,280 --> 00:24:18,720
In other plants it moves several times a week.

602
00:24:18,720 --> 00:24:21,520
The more it moves, the less useful a static plan becomes.

603
00:24:21,520 --> 00:24:23,440
People start reacting to visible cues,

604
00:24:23,440 --> 00:24:25,920
loud escalations, or whichever machine looks busiest,

605
00:24:25,920 --> 00:24:28,000
while the actual flow problem sits somewhere else.

606
00:24:28,000 --> 00:24:30,400
You can spot this pattern when local efficiency rises,

607
00:24:30,400 --> 00:24:32,640
but delivery performance stays unstable.

608
00:24:32,640 --> 00:24:35,280
Teams report strong output from individual areas,

609
00:24:35,280 --> 00:24:37,760
machines stay busy, OEE improves,

610
00:24:37,760 --> 00:24:39,840
yet orders still wait, lead time stretch,

611
00:24:39,840 --> 00:24:43,040
and planners can't give sales a confident answer about completion.

612
00:24:43,040 --> 00:24:45,440
The factory measures motion but not always flow.

613
00:24:45,440 --> 00:24:47,920
A planning tool can help connect those two views.

614
00:24:47,920 --> 00:24:50,320
It can model shared resources, downstream limits,

615
00:24:50,320 --> 00:24:51,680
and order dependencies,

616
00:24:51,680 --> 00:24:54,240
then test whether a proposed sequence feeds the constraint

617
00:24:54,240 --> 00:24:56,240
that currently controls delivery.

618
00:24:56,240 --> 00:24:58,000
That won't remove every cue.

619
00:24:58,000 --> 00:24:59,520
Some cues protect flow,

620
00:24:59,520 --> 00:25:02,960
but it can stop the plan from treating every available machine hour

621
00:25:02,960 --> 00:25:04,240
as equally useful.

622
00:25:04,240 --> 00:25:07,520
Because a production decision can't come from machine data alone.

623
00:25:07,520 --> 00:25:09,920
Machine data tells you what happened at one resource.

624
00:25:09,920 --> 00:25:12,240
Down time, rate, rejects, starters.

625
00:25:12,240 --> 00:25:14,400
To choose the next best sequence across the factory,

626
00:25:14,400 --> 00:25:16,800
you also need the relationship between that resource,

627
00:25:16,800 --> 00:25:19,360
the order, the next operation, the material state,

628
00:25:19,360 --> 00:25:21,200
and the customer date.

629
00:25:21,200 --> 00:25:23,360
The fifth warning sign, expedite work,

630
00:25:23,360 --> 00:25:25,040
becomes the normal workflow.

631
00:25:25,040 --> 00:25:26,240
Here's another warning sign,

632
00:25:26,240 --> 00:25:28,640
and it's one you can see from across the factory floor.

633
00:25:28,640 --> 00:25:30,560
Urgent work stops being the exception

634
00:25:30,560 --> 00:25:32,320
and becomes how the factory runs.

635
00:25:32,320 --> 00:25:35,520
Almost every manufacturer has to expedite an order sometimes.

636
00:25:35,520 --> 00:25:37,200
A customer with a line-down problem,

637
00:25:37,200 --> 00:25:39,040
a replacement part that needs to ship,

638
00:25:39,040 --> 00:25:42,000
a quality issue that created unexpected demand.

639
00:25:42,000 --> 00:25:44,640
Nobody needs a system to tell them that some situations

640
00:25:44,640 --> 00:25:46,080
deserve a fast response.

641
00:25:46,080 --> 00:25:49,520
The problem begins when every day brings a new "must run now" order,

642
00:25:49,520 --> 00:25:52,960
and nobody can explain what each priority override pushes out of the plan.

643
00:25:52,960 --> 00:25:55,920
Picture a shared production resource,

644
00:25:55,920 --> 00:25:57,600
a paint booth, a furnace,

645
00:25:57,600 --> 00:25:59,680
or a specialist machining operation.

646
00:25:59,680 --> 00:26:02,400
Sales escalates in order because the customer wants it sooner.

647
00:26:02,400 --> 00:26:03,600
The order moves to the front

648
00:26:03,600 --> 00:26:05,440
and production does its best to help,

649
00:26:05,440 --> 00:26:06,960
because the request sounds reasonable

650
00:26:06,960 --> 00:26:08,800
and the customer relationship matters.

651
00:26:08,800 --> 00:26:10,800
But that resource already has a sequence,

652
00:26:10,800 --> 00:26:12,720
one that may group products by material,

653
00:26:12,720 --> 00:26:16,160
color, tooling, or process conditions to protect another delivery date production

654
00:26:16,160 --> 00:26:17,680
has committed to four weeks.

655
00:26:17,680 --> 00:26:19,680
When the expedited order jumps ahead,

656
00:26:19,680 --> 00:26:21,600
people don't just gain time for that customer.

657
00:26:21,600 --> 00:26:23,280
They lose time through a changeover,

658
00:26:23,280 --> 00:26:24,320
disrupt the queue,

659
00:26:24,320 --> 00:26:27,520
and delay work that may become the next escalation tomorrow morning.

660
00:26:27,520 --> 00:26:29,200
Urgency moves through the factory.

661
00:26:29,200 --> 00:26:31,680
A planner can see that one order needs attention,

662
00:26:31,680 --> 00:26:34,480
but they may not quickly see the full cost of that decision,

663
00:26:34,480 --> 00:26:37,600
which orders move how much setup time the new sequence adds

664
00:26:37,600 --> 00:26:40,320
or whether the expedited job even has every part,

665
00:26:40,320 --> 00:26:43,440
approval, and resource it needs to finish.

666
00:26:43,440 --> 00:26:46,640
Or worse, it jumps the queue, consumes scarce capacity,

667
00:26:46,640 --> 00:26:48,560
and then waits at the next operation.

668
00:26:48,560 --> 00:26:50,160
That's why can we expedite this?

669
00:26:50,160 --> 00:26:51,440
That isn't the right question.

670
00:26:51,440 --> 00:26:52,640
A better question is,

671
00:26:52,640 --> 00:26:54,480
if we expedite this, what will it delay?

672
00:26:54,480 --> 00:26:56,400
And are we willing to accept that result?

673
00:26:56,400 --> 00:26:59,440
The first asks for effort, the second asks for a decision.

674
00:26:59,440 --> 00:27:01,280
That distinction changes the conversation

675
00:27:01,280 --> 00:27:03,200
between sales, planning, and production.

676
00:27:03,200 --> 00:27:05,920
Sales may still decide that a certain customer needs priority,

677
00:27:05,920 --> 00:27:08,240
but to say that's a valid commercial choice,

678
00:27:08,240 --> 00:27:11,280
but the decision should come with visible operational consequences.

679
00:27:11,280 --> 00:27:14,960
If moving order A ahead pushes order B past its promised date,

680
00:27:14,960 --> 00:27:16,480
somebody needs to own that trade-off

681
00:27:16,480 --> 00:27:18,640
before the floor gets another urgent message.

682
00:27:18,640 --> 00:27:21,360
Otherwise, the loudest voice becomes the scheduling method.

683
00:27:21,360 --> 00:27:23,760
You recognize this in the language people use.

684
00:27:23,760 --> 00:27:25,200
Just squeeze it in.

685
00:27:25,200 --> 00:27:26,880
Can we run this one first?

686
00:27:26,880 --> 00:27:28,320
This one really can't be late.

687
00:27:28,320 --> 00:27:29,600
Each request may be honest,

688
00:27:29,600 --> 00:27:31,920
but if the plan has no way to show displaced work,

689
00:27:31,920 --> 00:27:33,840
every request arrives as if it exists

690
00:27:33,840 --> 00:27:35,440
outside the rest of the factory.

691
00:27:35,440 --> 00:27:37,120
Production then carries the burden.

692
00:27:37,120 --> 00:27:38,880
Supervisors try to protect throughput

693
00:27:38,880 --> 00:27:40,800
while responding to calls from planning.

694
00:27:40,800 --> 00:27:42,400
Operators deal with broken sequences,

695
00:27:42,400 --> 00:27:44,000
more setups, and work that arrives

696
00:27:44,000 --> 00:27:45,360
without normal preparation.

697
00:27:45,360 --> 00:27:46,880
Planners negotiate priorities

698
00:27:46,880 --> 00:27:48,480
rather than test alternatives.

699
00:27:48,480 --> 00:27:50,800
And customers whose orders quietly slipped

700
00:27:50,800 --> 00:27:53,120
may hear about the delay only after the capacity

701
00:27:53,120 --> 00:27:54,720
has already been used elsewhere.

702
00:27:54,720 --> 00:27:57,120
Expediting can also hide deeper problems.

703
00:27:57,120 --> 00:27:59,920
If the same customer orders always need special treatment,

704
00:27:59,920 --> 00:28:02,480
maybe the promised date process is too optimistic.

705
00:28:02,480 --> 00:28:04,880
If shortages turn routine work into emergencies,

706
00:28:04,880 --> 00:28:07,360
material status may be entering planning too late.

707
00:28:07,360 --> 00:28:10,160
If every month ends with a rush through final operations,

708
00:28:10,160 --> 00:28:11,600
the plant may be releasing work

709
00:28:11,600 --> 00:28:12,480
without enough regard

710
00:28:12,480 --> 00:28:15,120
for where completion capacity actually sits.

711
00:28:15,120 --> 00:28:16,720
An expedite is a useful signal.

712
00:28:16,720 --> 00:28:18,720
It tells you that the normal planning process

713
00:28:18,720 --> 00:28:20,800
couldn't absorb a real business need.

714
00:28:20,800 --> 00:28:23,040
The right response isn't always too rejected.

715
00:28:23,040 --> 00:28:24,320
Sometimes it's too accepted,

716
00:28:24,320 --> 00:28:25,360
document the consequence,

717
00:28:25,360 --> 00:28:27,280
and revise the plan around that choice.

718
00:28:27,280 --> 00:28:28,960
But if nobody tracks the consequence,

719
00:28:28,960 --> 00:28:31,200
the exception process slowly replaces the plan

720
00:28:31,200 --> 00:28:32,720
that creates a strange situation.

721
00:28:32,720 --> 00:28:34,640
The official schedule still exists in ERP

722
00:28:34,640 --> 00:28:36,480
on a board or in a planning tool,

723
00:28:36,480 --> 00:28:39,120
but the actual schedule emerges through escalations,

724
00:28:39,120 --> 00:28:41,200
chat messages, hallway conversations,

725
00:28:41,200 --> 00:28:42,640
and whoever calls first.

726
00:28:42,640 --> 00:28:44,560
People work hard all day,

727
00:28:44,560 --> 00:28:46,640
but the factory loses a shared basis

728
00:28:46,640 --> 00:28:48,480
for deciding what matters most.

729
00:28:48,480 --> 00:28:51,200
Dedicated production planning software

730
00:28:51,200 --> 00:28:53,200
earns its place when it can test these changes

731
00:28:53,200 --> 00:28:54,960
against the full root of the order,

732
00:28:54,960 --> 00:28:56,880
showing whether the operation can move,

733
00:28:56,880 --> 00:28:58,320
which resource it needs,

734
00:28:58,320 --> 00:29:00,560
what work already claims that resource,

735
00:29:00,560 --> 00:29:02,800
and which completion dates change.

736
00:29:02,800 --> 00:29:05,680
It doesn't decide whether a customer deserves special treatment,

737
00:29:05,680 --> 00:29:07,200
people still own that decision,

738
00:29:07,200 --> 00:29:09,280
but it gives them a clearer view of the cost

739
00:29:09,280 --> 00:29:10,720
before they commit the factory.

740
00:29:10,720 --> 00:29:12,000
And to calculate that cost,

741
00:29:12,000 --> 00:29:15,120
the plan needs more than a list of orders and due dates.

742
00:29:15,120 --> 00:29:16,800
It needs to understand the dependencies

743
00:29:16,800 --> 00:29:18,480
that connect one operation to the next

744
00:29:18,480 --> 00:29:20,640
because an urgent order only looks simple

745
00:29:20,640 --> 00:29:22,400
until you follow it through the full routing.

746
00:29:22,400 --> 00:29:25,600
The sixth warning sign, too many dependencies.

747
00:29:25,600 --> 00:29:27,600
Live only in people's heads.

748
00:29:27,600 --> 00:29:29,280
Here's a warning sign you don't notice

749
00:29:29,280 --> 00:29:30,320
until someone leaves,

750
00:29:30,320 --> 00:29:31,280
takes a holiday,

751
00:29:31,280 --> 00:29:32,720
or moves to another shift.

752
00:29:32,720 --> 00:29:35,040
Your plan only works because certain people

753
00:29:35,040 --> 00:29:37,360
carry a detailed map of the factory in their heads.

754
00:29:37,360 --> 00:29:40,480
They know which product can use an alternate route,

755
00:29:40,480 --> 00:29:42,320
but only after quality approval,

756
00:29:42,320 --> 00:29:44,720
which customer won't accept a material substitute,

757
00:29:44,720 --> 00:29:47,200
even if engineering approved it for everyone else,

758
00:29:47,200 --> 00:29:49,920
and which tool technically fits two machines.

759
00:29:49,920 --> 00:29:52,400
But only one gives an acceptable surface finish

760
00:29:52,400 --> 00:29:53,920
on that part family.

761
00:29:53,920 --> 00:29:55,520
That knowledge comes from years of work,

762
00:29:55,520 --> 00:29:57,280
past failures, operator conversations,

763
00:29:57,280 --> 00:29:59,600
and solving the same awkward problems over and over.

764
00:29:59,600 --> 00:30:01,840
A good planner doesn't just see an order number,

765
00:30:01,840 --> 00:30:03,520
they see the exceptions around it,

766
00:30:03,520 --> 00:30:04,960
the risks nobody wrote down,

767
00:30:04,960 --> 00:30:08,160
and the small conditions that decide whether the plan will actually work.

768
00:30:08,160 --> 00:30:09,920
Now that kind of experience deserves respect,

769
00:30:09,920 --> 00:30:10,880
you don't want to lose it,

770
00:30:10,880 --> 00:30:12,960
but there's a real difference between expert judgment

771
00:30:12,960 --> 00:30:15,440
and a factory depending on undocumented rules.

772
00:30:15,440 --> 00:30:18,000
Expert judgment helps in unusual situations.

773
00:30:18,000 --> 00:30:21,120
Undocumented dependency means the normal situation works only

774
00:30:21,120 --> 00:30:23,200
because one person remembers how it runs.

775
00:30:23,200 --> 00:30:24,960
Consider a planner who gets a request

776
00:30:24,960 --> 00:30:27,440
to move an urgent job from one machine to another.

777
00:30:27,440 --> 00:30:28,960
On paper it looks easy.

778
00:30:28,960 --> 00:30:30,720
Both machines appear in the routing.

779
00:30:30,720 --> 00:30:32,160
The second machine has time,

780
00:30:32,160 --> 00:30:33,280
the material is available,

781
00:30:33,280 --> 00:30:35,760
and the order carries a near-term delivery date.

782
00:30:35,760 --> 00:30:37,600
Someone looking only at the system data might wonder

783
00:30:37,600 --> 00:30:39,600
why the planner hasn't moved it already.

784
00:30:39,600 --> 00:30:41,360
The planner knows the real reasons.

785
00:30:41,360 --> 00:30:43,200
The fixture is in use on another job.

786
00:30:43,200 --> 00:30:46,320
The next shift operator hasn't been approved for that customer's process,

787
00:30:46,320 --> 00:30:48,080
and even if both issues disappear,

788
00:30:48,080 --> 00:30:49,760
the alternate route adds a cleaning step

789
00:30:49,760 --> 00:30:52,640
because the first machine uses a different coolant.

790
00:30:52,640 --> 00:30:54,640
That extra step changes inspection timing,

791
00:30:54,640 --> 00:30:56,480
where another urgent order already waits.

792
00:30:56,480 --> 00:30:58,080
So the planner rejects the simple swap.

793
00:30:58,080 --> 00:30:59,360
That isn't resistance to change.

794
00:30:59,360 --> 00:31:00,960
It's planning context.

795
00:31:00,960 --> 00:31:03,680
The problem comes when the planner has to explain all of that

796
00:31:03,680 --> 00:31:06,080
through calls, notes, and memory every single time.

797
00:31:06,080 --> 00:31:09,440
And if they're absent, another planner might approve the move

798
00:31:09,440 --> 00:31:11,440
because the routing suggested should work,

799
00:31:11,440 --> 00:31:13,840
then production discovers the missing fixture.

800
00:31:13,840 --> 00:31:16,400
The order sits on the floor, people lose time,

801
00:31:16,400 --> 00:31:18,400
and trust in the plan drops another notch.

802
00:31:18,400 --> 00:31:20,080
You see the same issue with skills.

803
00:31:20,080 --> 00:31:21,200
A machine may have open time,

804
00:31:21,200 --> 00:31:22,880
but only a few people can set it up,

805
00:31:22,880 --> 00:31:24,320
program it, inspect the output,

806
00:31:24,320 --> 00:31:26,160
or sign off a regulated process.

807
00:31:26,160 --> 00:31:27,680
A calendar that counts machine hours

808
00:31:27,680 --> 00:31:29,760
without those qualification limits creates a schedule

809
00:31:29,760 --> 00:31:31,920
that looks feasible until the shift starts.

810
00:31:31,920 --> 00:31:33,520
The machine isn't the whole resource.

811
00:31:33,520 --> 00:31:35,520
The machine tool program material state

812
00:31:35,520 --> 00:31:37,280
and available skill together form the resource

813
00:31:37,280 --> 00:31:38,560
the order actually needs.

814
00:31:38,560 --> 00:31:40,080
That's more than a data detail.

815
00:31:40,080 --> 00:31:42,080
It affects how the plant responds to growth.

816
00:31:42,080 --> 00:31:44,640
When order volume rises more sites come online,

817
00:31:44,640 --> 00:31:46,480
or the business ads shifts,

818
00:31:46,480 --> 00:31:48,400
informal knowledge doesn't travel naturally.

819
00:31:48,400 --> 00:31:50,640
The planner can't stand beside every supervisor.

820
00:31:50,640 --> 00:31:52,880
The supervisor can't know every customer rule,

821
00:31:52,880 --> 00:31:55,600
and a new employee can't safely inherit years of exceptions

822
00:31:55,600 --> 00:31:56,880
from a folder of old emails.

823
00:31:57,680 --> 00:31:59,520
Retirement exposes this sharply,

824
00:31:59,520 --> 00:32:01,120
but you don't need to wait for retirement.

825
00:32:01,120 --> 00:32:02,560
A planner taking two weeks off

826
00:32:02,560 --> 00:32:04,880
can reveal which planning rules never enter the system.

827
00:32:04,880 --> 00:32:07,520
A shift handover can reveal that one team uses a workaround,

828
00:32:07,520 --> 00:32:08,720
another team doesn't know,

829
00:32:08,720 --> 00:32:10,720
and a new product can reveal that engineering

830
00:32:10,720 --> 00:32:12,240
defined a nominal route

831
00:32:12,240 --> 00:32:13,920
while production has already learned the route

832
00:32:13,920 --> 00:32:16,080
that works under real conditions.

833
00:32:16,080 --> 00:32:17,840
So production planning software can help.

834
00:32:17,840 --> 00:32:19,840
It gives you a place to model repeatable rules

835
00:32:19,840 --> 00:32:21,600
like approved alternate resources,

836
00:32:21,600 --> 00:32:23,120
two limits, setup families,

837
00:32:23,120 --> 00:32:24,320
operator qualifications,

838
00:32:24,320 --> 00:32:25,840
material substitution rules,

839
00:32:25,840 --> 00:32:28,080
or customer-specific process requirements.

840
00:32:28,080 --> 00:32:31,200
But the system doesn't need to capture every piece of human experience.

841
00:32:31,200 --> 00:32:33,440
If you try to turn all judgment into data,

842
00:32:33,440 --> 00:32:36,000
you usually end up with a model nobody can maintain.

843
00:32:36,000 --> 00:32:37,840
Instead, capture the rules that recur

844
00:32:37,840 --> 00:32:39,520
and that change planning decisions

845
00:32:39,520 --> 00:32:41,920
and keep the planner's judgment where it belongs.

846
00:32:41,920 --> 00:32:44,080
In exceptions, trade-offs, risk calls,

847
00:32:44,080 --> 00:32:45,760
and conditions, the model can't know.

848
00:32:45,760 --> 00:32:49,120
If a major customer relationship needs special treatment,

849
00:32:49,120 --> 00:32:50,560
a person should decide that.

850
00:32:50,560 --> 00:32:52,080
If an operator sees a problem,

851
00:32:52,080 --> 00:32:53,280
the data hasn't captured,

852
00:32:53,280 --> 00:32:55,280
the supervisor should challenge the plan.

853
00:32:55,280 --> 00:32:58,160
Software should support that conversation with clear assumptions,

854
00:32:58,160 --> 00:33:00,240
not silence it with a black box answer.

855
00:33:00,240 --> 00:33:03,040
Think of the planning model as shared operational memory.

856
00:33:03,040 --> 00:33:05,360
It records the rules the factory has already learned,

857
00:33:05,360 --> 00:33:08,240
so each planner doesn't need to rediscover them through disruption.

858
00:33:08,240 --> 00:33:10,560
When a rule changes, the team can discuss it,

859
00:33:10,560 --> 00:33:11,360
update it,

860
00:33:11,360 --> 00:33:13,280
and see where it affects future decisions.

861
00:33:13,280 --> 00:33:15,440
That gives experienced people more time

862
00:33:15,440 --> 00:33:17,600
for the cases that genuinely need experience.

863
00:33:17,600 --> 00:33:20,000
Once a factory needs to model those dependencies,

864
00:33:20,000 --> 00:33:23,040
it also needs to look closely at what the ERP can and can't carry.

865
00:33:23,680 --> 00:33:26,240
Why ERP planning often reaches its limit?

866
00:33:26,240 --> 00:33:28,720
Now, once you need those rules in a shared form,

867
00:33:28,720 --> 00:33:29,840
the fair question is,

868
00:33:29,840 --> 00:33:32,560
does the ERP already do production planning?

869
00:33:32,560 --> 00:33:33,760
Sometimes it does enough,

870
00:33:33,760 --> 00:33:35,200
and ERP should remain central

871
00:33:35,200 --> 00:33:37,680
because it holds the business facts that planning depends on.

872
00:33:37,680 --> 00:33:39,680
It knows customer orders, bills of material,

873
00:33:39,680 --> 00:33:42,240
inventory, purchasing status, production orders,

874
00:33:42,240 --> 00:33:44,000
and often the routines that describe

875
00:33:44,000 --> 00:33:46,000
how a product moves through the plan.

876
00:33:46,000 --> 00:33:48,880
It also connects production to cost, finance, and supply.

877
00:33:48,880 --> 00:33:51,840
You don't replace that role with a scheduling tool.

878
00:33:51,840 --> 00:33:54,800
ERP planning usually includes material requirements planning

879
00:33:54,800 --> 00:33:56,880
or MRP, which starts with demand.

880
00:33:56,880 --> 00:33:58,880
MRP looks at what customers have ordered,

881
00:33:58,880 --> 00:34:00,160
what forecasts suggest,

882
00:34:00,160 --> 00:34:01,680
what materials sit in stock,

883
00:34:01,680 --> 00:34:03,280
what suppliers need to deliver,

884
00:34:03,280 --> 00:34:05,840
and what components each finished product requires.

885
00:34:05,840 --> 00:34:08,400
From there, it proposes purchasing and production activity

886
00:34:08,400 --> 00:34:09,840
based on dates, lead times,

887
00:34:09,840 --> 00:34:11,120
and the product structure.

888
00:34:11,120 --> 00:34:12,240
That work matters.

889
00:34:12,240 --> 00:34:14,160
Without it, production planning starts

890
00:34:14,160 --> 00:34:16,880
with incomplete demand and uncertain material supply.

891
00:34:16,880 --> 00:34:19,840
An advanced scheduler can produce a very detailed sequence,

892
00:34:19,840 --> 00:34:21,840
but it can't help much if the order demand,

893
00:34:21,840 --> 00:34:24,800
bill of material, or purchase status coming into it is wrong.

894
00:34:24,800 --> 00:34:28,320
ERP gives the plant a commercial and material backbone.

895
00:34:28,320 --> 00:34:30,800
The issue appears when people ask ERP

896
00:34:30,800 --> 00:34:33,840
to answer a more detailed question than it was designed for.

897
00:34:33,840 --> 00:34:36,320
For example, you might need to decide which of 10 operations

898
00:34:36,320 --> 00:34:38,960
should run next Tuesday morning on a specific resource,

899
00:34:38,960 --> 00:34:41,360
given jobs already in progress, setup sequence,

900
00:34:41,360 --> 00:34:43,280
a maintenance window, current staffing,

901
00:34:43,280 --> 00:34:45,520
and the fact that one route is technically possible

902
00:34:45,520 --> 00:34:47,520
but undesirable for a certain customer.

903
00:34:48,000 --> 00:34:50,240
That question sits much closer to execution.

904
00:34:50,240 --> 00:34:52,960
Many ERP systems can hold work centers, calendars,

905
00:34:52,960 --> 00:34:54,640
rootings, and capacity data.

906
00:34:54,640 --> 00:34:56,640
They perform rough cut capacity planning,

907
00:34:56,640 --> 00:34:59,920
comparing expected load with available capacity at a broader level.

908
00:34:59,920 --> 00:35:01,920
They calculate dates from plant lead times,

909
00:35:01,920 --> 00:35:04,640
flag overloads, and help planners see demand building

910
00:35:04,640 --> 00:35:06,000
in a work center or period.

911
00:35:06,000 --> 00:35:07,120
That's useful planning,

912
00:35:07,120 --> 00:35:10,480
but rough capacity isn't the same as a feasible shop floor schedule.

913
00:35:10,480 --> 00:35:12,880
A weekly load picture might tell you that a work center

914
00:35:12,880 --> 00:35:14,800
faces more work than it's available hours

915
00:35:14,800 --> 00:35:16,560
but it can't decide the exact sequence

916
00:35:16,560 --> 00:35:18,000
that reduces setup loss,

917
00:35:18,000 --> 00:35:20,000
protects the due dates that matter most,

918
00:35:20,000 --> 00:35:21,600
respects every resource limit,

919
00:35:21,600 --> 00:35:24,000
and reacts well when the plan changes during the day.

920
00:35:24,000 --> 00:35:26,880
Think about the date in ERP gives an order.

921
00:35:26,880 --> 00:35:28,880
In many cases that date comes from a routing,

922
00:35:28,880 --> 00:35:30,320
standard operation times,

923
00:35:30,320 --> 00:35:32,880
queue time, transport time, and the calendar,

924
00:35:32,880 --> 00:35:34,640
reasonable planning assumptions that help create

925
00:35:34,640 --> 00:35:36,080
an initial expectation.

926
00:35:36,080 --> 00:35:38,080
Yet the date may not account for the actual queue

927
00:35:38,080 --> 00:35:39,360
in front of each operation,

928
00:35:39,360 --> 00:35:41,040
a tool already committed elsewhere,

929
00:35:41,040 --> 00:35:42,160
a temporary quality hold,

930
00:35:42,160 --> 00:35:44,560
or a shift where the only qualified person is absent.

931
00:35:45,200 --> 00:35:47,440
The ERP date tells you what the business expects,

932
00:35:47,440 --> 00:35:50,160
but it doesn't always tell you what the shop floor can still achieve.

933
00:35:50,160 --> 00:35:51,680
That isn't a floor in ERP.

934
00:35:51,680 --> 00:35:53,520
It's a question of architectural role.

935
00:35:53,520 --> 00:35:55,760
ERP needs to plan across purchasing inventory,

936
00:35:55,760 --> 00:35:57,440
customer demand, and the whole business,

937
00:35:57,440 --> 00:36:00,000
and it can't always carry every short term operational rule

938
00:36:00,000 --> 00:36:02,800
without becoming hard to maintain and slow to use.

939
00:36:02,800 --> 00:36:06,400
Here's what happens if you force every local production condition into ERP.

940
00:36:06,400 --> 00:36:07,760
People work around it anyway,

941
00:36:07,760 --> 00:36:10,400
they create side-spread sheets for real priorities,

942
00:36:10,400 --> 00:36:12,000
supervisors maintain local boards

943
00:36:12,000 --> 00:36:14,640
because the formal plan doesn't reflect current conditions,

944
00:36:14,640 --> 00:36:18,320
and plans keep private notes about alternate resources and setup rules.

945
00:36:18,320 --> 00:36:20,400
The ERP stays the system of record,

946
00:36:20,400 --> 00:36:22,240
but stops being the system people trust

947
00:36:22,240 --> 00:36:23,920
for the next production decision.

948
00:36:23,920 --> 00:36:25,200
That split creates trouble.

949
00:36:25,200 --> 00:36:27,200
The business and the floor drift apart.

950
00:36:27,200 --> 00:36:30,400
A better architecture keeps ERP in charge of what it does well.

951
00:36:30,400 --> 00:36:32,480
Owning demand, production orders,

952
00:36:32,480 --> 00:36:34,800
material structure, purchase signals,

953
00:36:34,800 --> 00:36:36,960
and the commercial facts around the work.

954
00:36:36,960 --> 00:36:39,520
A planning or scheduling layer can then use those facts,

955
00:36:39,520 --> 00:36:41,920
along with real capacity and production rules

956
00:36:41,920 --> 00:36:43,760
to test how the work can happen.

957
00:36:43,760 --> 00:36:46,640
The result should flow back when a constrained schedule shows an order

958
00:36:46,640 --> 00:36:48,160
can't meet its planned date

959
00:36:48,160 --> 00:36:50,160
that needs to inform customer communication,

960
00:36:50,160 --> 00:36:52,640
purchasing action, and management decisions.

961
00:36:52,640 --> 00:36:54,640
Planning shouldn't become another isolated system

962
00:36:54,640 --> 00:36:56,800
with its own private version of the future.

963
00:36:56,800 --> 00:36:58,800
It needs a controlled exchange with ERP

964
00:36:58,800 --> 00:37:00,880
where each system keeps a clear purpose.

965
00:37:00,880 --> 00:37:02,640
You can think of ERP as the system

966
00:37:02,640 --> 00:37:05,920
that asks what we need to make and what supply we need to support it.

967
00:37:05,920 --> 00:37:08,160
Detailed planning asks a different question.

968
00:37:08,160 --> 00:37:10,080
Given the factory as it is now,

969
00:37:10,080 --> 00:37:11,920
what sequence can we actually run?

970
00:37:11,920 --> 00:37:13,280
Both questions need answers.

971
00:37:13,280 --> 00:37:16,240
ERP can also struggle when conditions change quickly.

972
00:37:16,240 --> 00:37:18,880
MRP can react to revised demand and supply dates,

973
00:37:18,880 --> 00:37:21,040
but frequent recalculation doesn't automatically

974
00:37:21,040 --> 00:37:23,120
create a sensible sequence on the floor.

975
00:37:23,120 --> 00:37:24,800
A planner may see updated recommendations

976
00:37:24,800 --> 00:37:26,480
and still need to decide which operation

977
00:37:26,480 --> 00:37:27,840
gets scarce machine time first

978
00:37:27,840 --> 00:37:29,920
and what risk each choice creates.

979
00:37:29,920 --> 00:37:31,440
That gap becomes more visible

980
00:37:31,440 --> 00:37:35,520
when production starts feeding back real execution facts.

981
00:37:35,520 --> 00:37:38,480
The manufacturing execution system, or MES,

982
00:37:38,480 --> 00:37:40,560
can tell you that an operation started late,

983
00:37:40,560 --> 00:37:43,120
a quantity completed, a machine stopped,

984
00:37:43,120 --> 00:37:45,360
or quality placed material on hold.

985
00:37:45,360 --> 00:37:47,120
Those are facts the ERP needs,

986
00:37:47,120 --> 00:37:49,600
but facts alone don't select the next feasible plan.

987
00:37:49,600 --> 00:37:52,880
Why MES alone doesn't solve the planning problem?

988
00:37:52,880 --> 00:37:55,520
Here's what MES gives you that ERP never could on its own,

989
00:37:55,520 --> 00:37:57,520
a real view of what production is actually doing,

990
00:37:57,520 --> 00:37:59,200
not just what the plan said it should be doing.

991
00:37:59,200 --> 00:38:02,640
An MES tracks everything happening on the floor.

992
00:38:02,640 --> 00:38:04,880
It logs when an operation starts and ends,

993
00:38:04,880 --> 00:38:06,240
how much quantity move through,

994
00:38:06,240 --> 00:38:07,920
where material sits, who did the work,

995
00:38:07,920 --> 00:38:10,240
and where the quality released or held the result.

996
00:38:10,240 --> 00:38:11,280
In regulated plants,

997
00:38:11,280 --> 00:38:13,440
it also handles work instructions, traceability,

998
00:38:13,440 --> 00:38:15,120
and the records you need to prove compliance.

999
00:38:15,120 --> 00:38:18,240
And that kind of detail changes what your planning team can actually see.

1000
00:38:18,240 --> 00:38:21,200
Say your plan assumes an order started at 8 in the morning.

1001
00:38:21,200 --> 00:38:24,720
The MES shows it waited until 11 because the machine wasn't ready.

1002
00:38:24,720 --> 00:38:27,600
Now planning has a real fact to work with instead of a guess.

1003
00:38:27,600 --> 00:38:30,880
Same situation when an operation only finished part of its quantity

1004
00:38:30,880 --> 00:38:32,880
or a quality check blocked the rest.

1005
00:38:32,880 --> 00:38:35,920
The next plan needs to respond to what actually happened yesterday,

1006
00:38:35,920 --> 00:38:37,120
not what we assumed.

1007
00:38:37,120 --> 00:38:38,720
Now let's be clear about something.

1008
00:38:38,720 --> 00:38:40,800
Visibility and planning are not the same job,

1009
00:38:40,800 --> 00:38:42,560
even though people often act like they are.

1010
00:38:42,560 --> 00:38:45,680
Picture a supervisor scanning their MES on a busy afternoon.

1011
00:38:45,680 --> 00:38:47,280
They can see one machine stopped,

1012
00:38:47,280 --> 00:38:49,440
another order halfway through its operation,

1013
00:38:49,440 --> 00:38:51,040
a batch hanging at inspection.

1014
00:38:51,040 --> 00:38:52,720
That tells them where work is right now.

1015
00:38:52,720 --> 00:38:55,280
It doesn't automatically answer which of five waiting orders

1016
00:38:55,280 --> 00:38:56,720
should get the next available slot,

1017
00:38:56,720 --> 00:38:59,520
or which choice protects your most important customer commitments

1018
00:38:59,520 --> 00:39:01,600
without creating a bigger problem downstream.

1019
00:39:01,600 --> 00:39:02,960
That's the real distinction.

1020
00:39:02,960 --> 00:39:05,280
The MES reports where you are today.

1021
00:39:05,280 --> 00:39:07,440
Planning needs to figure out where you can go from there.

1022
00:39:07,440 --> 00:39:10,400
That different sounds small until the plant hits a real disruption.

1023
00:39:10,400 --> 00:39:13,680
Say the MES reports a machining operation finished two hours late.

1024
00:39:13,680 --> 00:39:15,440
The execution record might be perfect,

1025
00:39:15,440 --> 00:39:18,960
but the planner still has to figure out whether the next operation can start today,

1026
00:39:18,960 --> 00:39:21,040
whether the material and tooling are ready,

1027
00:39:21,040 --> 00:39:23,680
whether that resource is already claimed by another order,

1028
00:39:23,680 --> 00:39:27,040
and what happens to the other jobs if we shift the late order ahead.

1029
00:39:27,040 --> 00:39:29,680
Those are scheduling decisions, not execution questions.

1030
00:39:29,680 --> 00:39:31,920
An MES often includes dispatching rules,

1031
00:39:31,920 --> 00:39:36,400
queue views, and priority lists useful when a work area needs to know what to run next.

1032
00:39:36,400 --> 00:39:38,400
But a dispatch list is only trustworthy

1033
00:39:38,400 --> 00:39:41,760
if the logic behind it sees enough of the wider production context.

1034
00:39:41,760 --> 00:39:44,720
Otherwise, the MES might tell the floor to run order 4-11T next

1035
00:39:44,720 --> 00:39:46,400
because it has the earliest due date.

1036
00:39:46,400 --> 00:39:49,200
Meanwhile, the planner knows order 4-23 should go first

1037
00:39:49,200 --> 00:39:52,480
because it feeds a constrained final assembly step tomorrow morning.

1038
00:39:52,480 --> 00:39:54,240
The execution system sees the status,

1039
00:39:54,240 --> 00:39:57,040
planning sees the consequence across the whole route.

1040
00:39:57,040 --> 00:39:59,360
Both views belong in the same operating loop.

1041
00:39:59,360 --> 00:40:02,000
The plan gives you a proposed sequence and expected dates.

1042
00:40:02,000 --> 00:40:04,000
The MES tells you what actually happened.

1043
00:40:04,000 --> 00:40:05,840
A disruption changes the choices,

1044
00:40:05,840 --> 00:40:08,000
planning produces a revised option,

1045
00:40:08,000 --> 00:40:10,000
and production gets an approved priority.

1046
00:40:10,000 --> 00:40:13,360
Without that loop, the schedule becomes a document created in the morning,

1047
00:40:13,360 --> 00:40:17,040
and the MES becomes a file that explains why the document was wrong by noon.

1048
00:40:17,040 --> 00:40:19,600
That feedback loop needs more than a nightly data transfer

1049
00:40:19,600 --> 00:40:22,160
if the plant changes quickly and most plants do.

1050
00:40:22,160 --> 00:40:24,800
For some planning decisions, a daily update is plenty.

1051
00:40:24,800 --> 00:40:28,400
A rough weekly capacity view doesn't need second by second machine signals.

1052
00:40:28,400 --> 00:40:31,440
But for near-term sequencing, the system needs current facts.

1053
00:40:31,440 --> 00:40:35,440
Completed work, downtime, quality status, material release.

1054
00:40:35,440 --> 00:40:38,400
At exactly the points where those facts change the next decision,

1055
00:40:38,400 --> 00:40:41,760
the practical question isn't whether you need real-time data everywhere.

1056
00:40:41,760 --> 00:40:44,400
It's which execution facts need to reach planning

1057
00:40:44,400 --> 00:40:46,480
fast enough to actually change a decision.

1058
00:40:46,480 --> 00:40:49,760
A machine state matters immediately if it's a constrained resource.

1059
00:40:49,760 --> 00:40:53,040
A completed operation matters when it releases work into a bottleneck.

1060
00:40:53,040 --> 00:40:56,960
A quality hold matters before the planner makes any promises to a customer.

1061
00:40:56,960 --> 00:41:00,800
This also explains why an MES project can improve shop floor discipline

1062
00:41:00,800 --> 00:41:02,560
without reducing planning chaos.

1063
00:41:02,560 --> 00:41:05,760
The plant may record better execution data, traceability improves,

1064
00:41:05,760 --> 00:41:07,760
and supervisors see more current status.

1065
00:41:07,760 --> 00:41:10,880
But planners still reconcile priorities in calls and spreadsheets

1066
00:41:10,880 --> 00:41:14,160
because no planning layer turns those facts into tested alternatives

1067
00:41:14,160 --> 00:41:17,520
under real capacity, sequence, and due date rules.

1068
00:41:17,520 --> 00:41:18,720
So here's where we land.

1069
00:41:18,720 --> 00:41:21,920
MES should feed planning and planning should respond to MES.

1070
00:41:21,920 --> 00:41:24,160
Neither system should pretend to own the other's job.

1071
00:41:24,160 --> 00:41:26,640
The MES records and controls execution.

1072
00:41:26,640 --> 00:41:29,120
A production planning system evaluates future options

1073
00:41:29,120 --> 00:41:32,160
and decides how to use available capacity to support demand.

1074
00:41:32,160 --> 00:41:36,000
ERP continues to own the commercial and material backbone around both.

1075
00:41:36,000 --> 00:41:38,800
Once that boundary is clear, the next practical question

1076
00:41:38,800 --> 00:41:42,240
becomes what dedicated production planning software should actually calculate

1077
00:41:42,240 --> 00:41:44,080
beyond just a better looking schedule.

1078
00:41:44,080 --> 00:41:46,720
What dedicated production planning software does?

1079
00:41:46,720 --> 00:41:49,520
Dedicated production planning software takes the planning facts

1080
00:41:49,520 --> 00:41:52,560
you already hold across systems and turns them into possible plans

1081
00:41:52,560 --> 00:41:54,400
that respect the rules of the factory

1082
00:41:54,400 --> 00:41:56,800
rather than just organizing orders by due date.

1083
00:41:56,800 --> 00:41:58,960
That sounds obvious, but the keyword is possible.

1084
00:41:58,960 --> 00:42:02,240
A list of orders sorted by due date isn't automatically feasible.

1085
00:42:02,240 --> 00:42:06,000
A schedule is only useful when it tests whether the work can actually happen

1086
00:42:06,000 --> 00:42:09,760
with the resources, time, material, and operating rules you have.

1087
00:42:09,760 --> 00:42:11,440
Think about an order entering planning.

1088
00:42:11,440 --> 00:42:14,640
It brings demand from the business side, quantity, requested date,

1089
00:42:14,640 --> 00:42:16,960
customer priority, product definition.

1090
00:42:16,960 --> 00:42:19,360
Its routing tells you what operations it needs.

1091
00:42:19,360 --> 00:42:22,480
Then the planning model asks which resources can run those operations,

1092
00:42:22,480 --> 00:42:24,480
how long they take, what constraints apply,

1093
00:42:24,480 --> 00:42:26,320
and what work is already there.

1094
00:42:26,320 --> 00:42:30,160
From there the system can produce options that are actually grounded in reality.

1095
00:42:30,160 --> 00:42:34,160
It tests finite capacity, meaning it won't put more work into a shift

1096
00:42:34,160 --> 00:42:35,760
than the resource can physically handle.

1097
00:42:35,760 --> 00:42:38,960
It checks sequence rules so a product family can run after similar work

1098
00:42:38,960 --> 00:42:40,560
to reduce unnecessary setup.

1099
00:42:40,560 --> 00:42:43,840
It considers approved alternate resources where they genuinely apply.

1100
00:42:43,840 --> 00:42:46,640
And it accounts for material readiness when that status determines

1101
00:42:46,640 --> 00:42:48,240
whether a job can move forward.

1102
00:42:48,240 --> 00:42:50,480
None of that means every rule needs automation.

1103
00:42:50,480 --> 00:42:52,240
Some decisions belong to people.

1104
00:42:52,240 --> 00:42:55,200
Customer relationships, commercial judgment, safety,

1105
00:42:55,200 --> 00:42:56,400
unstable conditions.

1106
00:42:56,400 --> 00:42:57,920
The planning engine has a different job.

1107
00:42:57,920 --> 00:43:00,000
It handles the repetitive calculation work

1108
00:43:00,000 --> 00:43:02,400
so the planner can see what each choice costs

1109
00:43:02,400 --> 00:43:06,160
before committing to it instead of discovering the damage after work has already started.

1110
00:43:06,160 --> 00:43:08,160
Let's put that into a factory scenario.

1111
00:43:08,160 --> 00:43:11,280
A planner needs to decide whether to move a late order ahead of two jobs

1112
00:43:11,280 --> 00:43:14,000
already scheduled on a constrained finishing resource.

1113
00:43:14,000 --> 00:43:16,560
One option protects the late order's delivery date.

1114
00:43:16,560 --> 00:43:19,600
Another protects a larger customer shipment later in the week.

1115
00:43:19,600 --> 00:43:23,200
A third option sends part of the work to an approved alternate resource

1116
00:43:23,200 --> 00:43:26,720
but that route adds time and creates risk around available labor.

1117
00:43:26,720 --> 00:43:29,680
A proper planning system doesn't just place the late order at the top

1118
00:43:29,680 --> 00:43:31,920
it calculates the impact of each option.

1119
00:43:31,920 --> 00:43:35,120
When each affected order would finish where overload appears

1120
00:43:35,120 --> 00:43:38,960
which setup sequence changes whether the alternate path still works.

1121
00:43:38,960 --> 00:43:42,240
Then the planner and production lead can make the trade-off with their eyes open

1122
00:43:42,240 --> 00:43:43,680
instead of hoping for the best.

1123
00:43:43,680 --> 00:43:46,160
That is decision support, not scheduling theatre.

1124
00:43:46,160 --> 00:43:49,360
The software may use different methods depending on the problem.

1125
00:43:49,360 --> 00:43:52,080
Advanced planning and scheduling, APS for short,

1126
00:43:52,080 --> 00:43:55,200
usually combines capacity aware planning with detailed scheduling.

1127
00:43:55,200 --> 00:43:58,480
Constraint based scheduling applies hard limits like a required tool,

1128
00:43:58,480 --> 00:44:02,000
a limited resource calendar or a mandatory operation sequence.

1129
00:44:02,000 --> 00:44:06,240
Optimization compares many feasible plans against goals the factory has defined.

1130
00:44:06,240 --> 00:44:10,480
Reducing lateness, cutting setup time, protecting a class of orders.

1131
00:44:10,480 --> 00:44:12,000
Simulation does something different.

1132
00:44:12,000 --> 00:44:15,200
It lets the team test assumptions before changing the live plan.

1133
00:44:15,200 --> 00:44:17,200
Say a supplier delivery slips by two days.

1134
00:44:17,200 --> 00:44:19,360
What if a weekend shift becomes available?

1135
00:44:19,360 --> 00:44:21,840
What if a resource loses half a shift for maintenance?

1136
00:44:21,840 --> 00:44:24,720
Simulation doesn't promise the future will happen that way.

1137
00:44:24,720 --> 00:44:27,120
It gives people a structured way to test exposure

1138
00:44:27,120 --> 00:44:29,600
before the disruption becomes urgent.

1139
00:44:29,600 --> 00:44:33,200
These methods can overlap in one product, but they aren't interchangeable.

1140
00:44:33,200 --> 00:44:35,200
A scheduler creates a feasible sequence

1141
00:44:35,200 --> 00:44:37,520
without finding the mathematically best one.

1142
00:44:37,520 --> 00:44:39,600
An optimizer searches for a better answer,

1143
00:44:39,600 --> 00:44:42,160
but only against the goals and rules it receives.

1144
00:44:42,160 --> 00:44:44,480
A simulation model may show risk across scenarios

1145
00:44:44,480 --> 00:44:46,800
without issuing the daily dispatch order.

1146
00:44:46,800 --> 00:44:49,120
Vendors sometimes throw all these words on one slide

1147
00:44:49,120 --> 00:44:50,800
and let the listener sort it out later.

1148
00:44:50,800 --> 00:44:52,400
That's a generous way to describe it.

1149
00:44:52,400 --> 00:44:53,760
The practical test is simpler.

1150
00:44:53,760 --> 00:44:56,400
Ask yourself what decision this tool needs to improve.

1151
00:44:56,400 --> 00:44:59,280
If planners struggle to find a feasible next sequence,

1152
00:44:59,280 --> 00:45:01,040
scheduling logic matters.

1153
00:45:01,040 --> 00:45:03,200
If they need to choose between competing service,

1154
00:45:03,200 --> 00:45:05,840
cost and setup goals, optimization may help.

1155
00:45:05,840 --> 00:45:09,280
If management needs to understand the effect of likely disruption,

1156
00:45:09,280 --> 00:45:11,760
simulation provides a safer way to explore options.

1157
00:45:11,760 --> 00:45:13,760
The planning engine sits behind all of this.

1158
00:45:13,760 --> 00:45:17,120
It isn't just a calendar with bars, colors, and drag and drop controls.

1159
00:45:17,120 --> 00:45:19,040
Those views help people discuss a plan,

1160
00:45:19,040 --> 00:45:20,640
but the view doesn't create the logic.

1161
00:45:20,640 --> 00:45:22,800
A gant chart with poor resource rules

1162
00:45:22,800 --> 00:45:26,160
just organizes confusion into neat horizontal lines.

1163
00:45:26,160 --> 00:45:28,320
Good planning software creates a decision model.

1164
00:45:28,320 --> 00:45:30,080
That model connects demand to operations,

1165
00:45:30,080 --> 00:45:32,320
operations to resources, resources to constraints

1166
00:45:32,320 --> 00:45:33,920
and constraints to completion dates.

1167
00:45:33,920 --> 00:45:36,960
When someone changes a priority, adds capacity, shifts an order

1168
00:45:36,960 --> 00:45:38,400
or reports a disruption,

1169
00:45:38,400 --> 00:45:40,480
the model recalculates the affected choices

1170
00:45:40,480 --> 00:45:41,680
and shows you the result.

1171
00:45:41,680 --> 00:45:43,680
The system should also explain its answer.

1172
00:45:43,680 --> 00:45:46,480
If it pushes an order later, the planner needs to know why.

1173
00:45:46,480 --> 00:45:50,000
Resource limit material block another order priority, sequence rules.

1174
00:45:50,000 --> 00:45:52,960
Without that explanation, people won't trust the schedule and they shouldn't.

1175
00:45:52,960 --> 00:45:56,000
A plan that can't explain itself becomes another source of arguments

1176
00:45:56,000 --> 00:45:57,680
rather than a tool for good decisions.

1177
00:45:57,680 --> 00:45:59,680
So dedicated planning software earns its place

1178
00:45:59,680 --> 00:46:02,080
when it helps people compare feasible options faster

1179
00:46:02,080 --> 00:46:03,280
and with less guesswork.

1180
00:46:03,280 --> 00:46:06,240
But feasibility depends on what the system knows about the factory.

1181
00:46:06,240 --> 00:46:09,360
If it's rooting, resource rules, and operating constraints

1182
00:46:09,360 --> 00:46:11,360
don't match the work people actually do.

1183
00:46:11,360 --> 00:46:13,360
The software can calculate very quickly

1184
00:46:13,360 --> 00:46:15,360
and still send the team in the wrong direction.

1185
00:46:15,360 --> 00:46:16,880
That's not a software problem.

1186
00:46:16,880 --> 00:46:19,200
That's a modeling problem that no algorithm can fix.

1187
00:46:19,200 --> 00:46:23,360
The data model question, most buying projects skip.

1188
00:46:23,360 --> 00:46:25,760
Here's the problem most manufacturers don't talk about

1189
00:46:25,760 --> 00:46:27,520
before buying planning software.

1190
00:46:27,520 --> 00:46:29,520
You get excited about screens and features,

1191
00:46:29,520 --> 00:46:32,480
watch a demo where sample orders slide across a neat calendar

1192
00:46:32,480 --> 00:46:33,840
and the plan looks solid.

1193
00:46:33,840 --> 00:46:35,440
Then the project hits real data.

1194
00:46:35,440 --> 00:46:37,280
The routing describes an ideal process

1195
00:46:37,280 --> 00:46:40,000
but production follows a different one when the week gets tough.

1196
00:46:40,000 --> 00:46:42,800
That gap decides whether the software actually helps

1197
00:46:42,800 --> 00:46:45,680
or just creates another argument to have in meetings.

1198
00:46:45,680 --> 00:46:47,200
So let's think about the planning model

1199
00:46:47,200 --> 00:46:48,800
through three connected ideas,

1200
00:46:48,800 --> 00:46:50,640
product, process, and resource.

1201
00:46:50,640 --> 00:46:52,560
Product means what the order requires.

1202
00:46:52,560 --> 00:46:54,800
The part or assembly, its revision, quantity,

1203
00:46:54,800 --> 00:46:58,320
quality specs, and sometimes customer-specific conditions.

1204
00:46:58,320 --> 00:47:01,120
Process is how that product moves through the factory.

1205
00:47:01,120 --> 00:47:03,200
A routing may list cutting, machining,

1206
00:47:03,200 --> 00:47:05,200
inspection, assembly, and packing.

1207
00:47:05,200 --> 00:47:08,160
But a useful planning model needs more than operation names.

1208
00:47:08,160 --> 00:47:10,880
It needs to know which steps must happen in sequence,

1209
00:47:10,880 --> 00:47:14,560
which work can overlap, which operation can use an alternate path

1210
00:47:14,560 --> 00:47:16,960
and where a batch can split or has to stay together.

1211
00:47:16,960 --> 00:47:18,000
Then there's the resource.

1212
00:47:18,000 --> 00:47:20,000
A resource isn't always just a machine.

1213
00:47:20,000 --> 00:47:22,000
It can be a machine with a certain tool,

1214
00:47:22,000 --> 00:47:24,160
a test cell with a qualified operator,

1215
00:47:24,160 --> 00:47:25,680
a furnace with a loading rule,

1216
00:47:25,680 --> 00:47:28,720
or a packing area that only handles certain product types.

1217
00:47:28,720 --> 00:47:31,600
The plan becomes believable when the model

1218
00:47:31,600 --> 00:47:34,240
connects a product requirement to a real process

1219
00:47:34,240 --> 00:47:36,720
and then to the resources that can perform that process

1220
00:47:36,720 --> 00:47:37,680
under current rules.

1221
00:47:37,680 --> 00:47:39,040
That sounds basic, but in practice,

1222
00:47:39,040 --> 00:47:40,640
those facts sit in different places.

1223
00:47:40,640 --> 00:47:42,560
ERP may own the product master,

1224
00:47:42,560 --> 00:47:45,520
builds of material production orders, and formal routines.

1225
00:47:45,520 --> 00:47:48,080
MES records actual execution and status,

1226
00:47:48,080 --> 00:47:50,960
maintenance holds planned downtime and repair information,

1227
00:47:50,960 --> 00:47:54,960
quality manages inspection plans, holds, approvals, and non-conformance.

1228
00:47:54,960 --> 00:47:57,680
Engineering controls revisions and technical instructions.

1229
00:47:57,680 --> 00:48:00,640
No single team owns the full planning picture by default,

1230
00:48:00,640 --> 00:48:01,760
and that isn't a failure.

1231
00:48:01,760 --> 00:48:05,440
Each system and team has a reason for owning its part.

1232
00:48:05,440 --> 00:48:07,840
The problem appears when a planning project assumes

1233
00:48:07,840 --> 00:48:10,720
that collecting extracts from each system automatically creates

1234
00:48:10,720 --> 00:48:12,400
shared meaning, which it doesn't.

1235
00:48:12,400 --> 00:48:15,840
A resource code in ERP may refer to a broad work center.

1236
00:48:15,840 --> 00:48:18,400
The same code in MES may point to a physical machine.

1237
00:48:18,400 --> 00:48:21,120
Maintenance tracks that machine under an asset tag.

1238
00:48:21,120 --> 00:48:23,600
Production calls it by a nickname that everyone understands

1239
00:48:23,600 --> 00:48:25,280
until someone new joins the team.

1240
00:48:25,280 --> 00:48:27,840
The software can't infer those relationships safely.

1241
00:48:27,840 --> 00:48:29,600
Let's make this concrete.

1242
00:48:29,600 --> 00:48:32,640
Imagine the ERP routing sensor product through machine group 12.

1243
00:48:32,640 --> 00:48:35,840
On the floor, planners know most orders use one specific machine

1244
00:48:35,840 --> 00:48:37,760
because it carries the preferred tooling.

1245
00:48:37,760 --> 00:48:41,040
A second machine can run the work, but only for certain material grades,

1246
00:48:41,040 --> 00:48:44,000
and quantity requires a different inspection step afterward.

1247
00:48:44,000 --> 00:48:45,760
None of that appears in the formal routing.

1248
00:48:45,760 --> 00:48:47,440
The routing isn't necessarily wrong.

1249
00:48:47,440 --> 00:48:49,760
It captures the standard method that helps ERP

1250
00:48:49,760 --> 00:48:51,520
create orders and plan demand,

1251
00:48:51,520 --> 00:48:54,400
but production has learned practical conditions over time,

1252
00:48:54,400 --> 00:48:56,960
and those conditions now shape daily decisions.

1253
00:48:56,960 --> 00:48:59,680
When planning software receives only the formal routing,

1254
00:48:59,680 --> 00:49:01,680
it may schedule work onto the alternate machine

1255
00:49:01,680 --> 00:49:03,200
because the calendar looks open.

1256
00:49:03,200 --> 00:49:05,920
The schedule then fails for a perfectly predictable reason.

1257
00:49:05,920 --> 00:49:08,320
That's why data quality and planning is operational governance,

1258
00:49:08,320 --> 00:49:10,000
not an IT cleanup exercise.

1259
00:49:10,000 --> 00:49:13,360
Somebody needs to decide which route reflects approved practice,

1260
00:49:13,360 --> 00:49:15,600
who can change an alternate resource rule

1261
00:49:15,600 --> 00:49:17,520
when a setup condition matters,

1262
00:49:17,520 --> 00:49:21,200
and how the plant treats exceptions that recur often enough to enter the model.

1263
00:49:21,200 --> 00:49:23,680
That work needs planners, supervisors,

1264
00:49:23,680 --> 00:49:26,400
engineering, quality, and maintenance in the same discussion.

1265
00:49:26,400 --> 00:49:28,800
I'd can build the integration and data teams can help create

1266
00:49:28,800 --> 00:49:30,480
a common structure and track lineage,

1267
00:49:30,480 --> 00:49:33,840
but they can't decide whether an operation really takes 40 minutes,

1268
00:49:33,840 --> 00:49:35,840
whether a route requires a skilled operator,

1269
00:49:35,840 --> 00:49:38,400
or whether a quality hold blocks a downstream step.

1270
00:49:38,400 --> 00:49:40,320
Those answers belong close to the work.

1271
00:49:40,320 --> 00:49:43,680
Now, you also don't need to model every edge case before you begin.

1272
00:49:43,680 --> 00:49:45,440
Trying to capture every possible exception

1273
00:49:45,440 --> 00:49:47,760
creates a planning model so hard to maintain

1274
00:49:47,760 --> 00:49:49,120
that people stop trusting it.

1275
00:49:49,120 --> 00:49:51,760
Start with the relationships that repeatedly affect the decision

1276
00:49:51,760 --> 00:49:52,560
you want to improve,

1277
00:49:52,560 --> 00:49:53,920
if the same tool conflict,

1278
00:49:53,920 --> 00:49:55,360
machine restriction setup rule,

1279
00:49:55,360 --> 00:49:57,840
or quality condition causes disruption each week.

1280
00:49:57,840 --> 00:49:58,960
Model that first,

1281
00:49:58,960 --> 00:50:01,280
use the model as a living operational agreement.

1282
00:50:01,280 --> 00:50:02,560
When a rule changes,

1283
00:50:02,560 --> 00:50:05,840
don't bury it in a planner's private note or supervises memory.

1284
00:50:05,840 --> 00:50:07,600
Review it, decide who owns it,

1285
00:50:07,600 --> 00:50:10,000
and update the model where it affects the plan.

1286
00:50:10,000 --> 00:50:13,520
Over time, the planning system gets closer to how the factory actually runs

1287
00:50:13,520 --> 00:50:15,600
rather than how someone expected it to run

1288
00:50:15,600 --> 00:50:17,440
when the routing first entered ERP.

1289
00:50:17,440 --> 00:50:20,000
The purchasing question then changes.

1290
00:50:20,000 --> 00:50:23,440
Instead of asking which planning tool has the most impressive demo,

1291
00:50:23,440 --> 00:50:27,440
ask whether your product, process, and resource relationships contain enough shared truth

1292
00:50:27,440 --> 00:50:29,120
for one real planning decision.

1293
00:50:29,120 --> 00:50:30,640
That's a much less glamorous question,

1294
00:50:30,640 --> 00:50:33,920
but it's the one that prevents a very expensive calendar from scheduling work

1295
00:50:33,920 --> 00:50:36,160
that production already knows cannot happen.

1296
00:50:36,160 --> 00:50:38,240
Minimum planning data before you buy.

1297
00:50:38,240 --> 00:50:40,880
You don't need perfect factory data before you start,

1298
00:50:40,880 --> 00:50:43,760
but you do need enough trusted data to answer one planning question

1299
00:50:43,760 --> 00:50:45,520
without building the answer on guesses.

1300
00:50:45,520 --> 00:50:47,040
Start with current demand.

1301
00:50:47,040 --> 00:50:49,120
The open production orders you intend to plan,

1302
00:50:49,120 --> 00:50:52,000
their quantities require dates and their current stage.

1303
00:50:52,000 --> 00:50:53,680
If an order has already started,

1304
00:50:53,680 --> 00:50:55,680
the planning model needs to know what remains,

1305
00:50:55,680 --> 00:50:57,920
not treat the whole order as untouched.

1306
00:50:57,920 --> 00:50:59,040
Next comes the root.

1307
00:50:59,040 --> 00:51:01,520
For the product family or plant area in scope,

1308
00:51:01,520 --> 00:51:03,440
you need the operations the work passes through,

1309
00:51:03,440 --> 00:51:06,080
their normal durations and the order in which they occur.

1310
00:51:06,080 --> 00:51:08,080
Those times won't be exact every day,

1311
00:51:08,080 --> 00:51:09,360
and they don't need to be,

1312
00:51:09,360 --> 00:51:12,880
but they need to be realistic enough that a plan can distinguish a short operation

1313
00:51:12,880 --> 00:51:16,320
from one that consumes most of a shift on a constrained resource.

1314
00:51:16,320 --> 00:51:18,000
Resource calendars matter just as much.

1315
00:51:18,000 --> 00:51:21,360
A machine calendar should reflect when that resource can actually run,

1316
00:51:21,360 --> 00:51:23,280
include shifts, planned shutdowns,

1317
00:51:23,280 --> 00:51:24,720
and known maintenance windows,

1318
00:51:24,720 --> 00:51:26,320
where they affect the decision.

1319
00:51:26,320 --> 00:51:28,400
If a resource needs a skilled operator,

1320
00:51:28,400 --> 00:51:31,200
and that skill limits when the work can run,

1321
00:51:31,200 --> 00:51:32,720
capture that condition too,

1322
00:51:32,720 --> 00:51:35,840
an apparently free machine doesn't help if nobody on shift can run the job.

1323
00:51:35,840 --> 00:51:37,760
Material status belongs in the model

1324
00:51:37,760 --> 00:51:40,320
when material regularly controls release decisions.

1325
00:51:40,320 --> 00:51:43,680
You don't need a perfect live feed for every bolt, label and consumable,

1326
00:51:43,680 --> 00:51:47,360
but if a shortage of a purchased component routinely stops final assembly,

1327
00:51:47,360 --> 00:51:49,840
then the planning decision needs a usable indication

1328
00:51:49,840 --> 00:51:51,840
of whether that component is available,

1329
00:51:51,840 --> 00:51:54,400
expected, held, or uncertain.

1330
00:51:54,400 --> 00:51:57,840
Otherwise, the scheduler can create a neat plan for work that cannot finish.

1331
00:51:57,840 --> 00:51:59,600
Same goes for alternate resources.

1332
00:51:59,600 --> 00:52:01,760
Only capture alternatives that production has approved

1333
00:52:01,760 --> 00:52:03,280
and can use under known conditions.

1334
00:52:03,280 --> 00:52:05,680
Don't list every machine that might possibly run a job

1335
00:52:05,680 --> 00:52:07,360
after a week of engineering work,

1336
00:52:07,360 --> 00:52:09,680
special tooling, and a lot of cross-fingers.

1337
00:52:09,680 --> 00:52:12,480
A planning rule needs to describe the choices the factory can make

1338
00:52:12,480 --> 00:52:14,000
within the horizon you're planning.

1339
00:52:14,000 --> 00:52:15,680
Setup logic should enter early,

1340
00:52:15,680 --> 00:52:17,680
where it creates recurring disruption.

1341
00:52:17,680 --> 00:52:21,200
Say a finishing cell needs a long clean down between two material types.

1342
00:52:21,200 --> 00:52:22,640
If the plan ignores that condition,

1343
00:52:22,640 --> 00:52:25,280
it may create a sequence that looks productive in the model

1344
00:52:25,280 --> 00:52:27,840
and collapses as soon as the team starts changing over.

1345
00:52:27,840 --> 00:52:30,000
You don't need to model every preference on day one,

1346
00:52:30,000 --> 00:52:32,640
but recurring setup loss belongs in the first version

1347
00:52:32,640 --> 00:52:34,160
if it changes daily decisions.

1348
00:52:34,160 --> 00:52:36,320
Here's a practical way to think about scope.

1349
00:52:36,320 --> 00:52:38,960
Choose the constraints that repeatedly cause late orders,

1350
00:52:38,960 --> 00:52:42,240
overtime, unnecessary work in progress, or planning arguments.

1351
00:52:42,240 --> 00:52:45,440
Then model those constraints well enough to test one bounded decision.

1352
00:52:45,440 --> 00:52:48,720
For example, can the plant produce a feasible daily sequence

1353
00:52:48,720 --> 00:52:50,560
for the constrained machining group

1354
00:52:50,560 --> 00:52:53,600
using open orders, approved routes, machine calendars,

1355
00:52:53,600 --> 00:52:56,320
setup families, and material readiness for the parts

1356
00:52:56,320 --> 00:52:58,000
that feed final assembly?

1357
00:52:58,000 --> 00:52:59,680
That's a real first use case.

1358
00:52:59,680 --> 00:53:02,240
It's much better than trying to model the full factory,

1359
00:53:02,240 --> 00:53:04,720
every product, every exception, and every data source

1360
00:53:04,720 --> 00:53:06,640
before anyone has used the planning system

1361
00:53:06,640 --> 00:53:08,400
to solve a live problem.

1362
00:53:08,400 --> 00:53:11,040
Perfection can become a very polite form of delay,

1363
00:53:11,040 --> 00:53:13,360
especially when each department waits for another department

1364
00:53:13,360 --> 00:53:14,560
to clean up something first.

1365
00:53:14,560 --> 00:53:16,320
Planning data improves through use,

1366
00:53:16,320 --> 00:53:18,240
when planners compare the proposed schedule

1367
00:53:18,240 --> 00:53:20,960
with what production knows, gaps become visible.

1368
00:53:20,960 --> 00:53:22,800
An operation time may need correction,

1369
00:53:22,800 --> 00:53:25,200
a resource rule may belong at a more detailed level,

1370
00:53:25,200 --> 00:53:28,240
or a material status may arrive too late to support the decision.

1371
00:53:28,240 --> 00:53:31,040
Those are useful findings because they connect data work

1372
00:53:31,040 --> 00:53:32,880
to an operational consequence.

1373
00:53:32,880 --> 00:53:36,000
Still, don't confuse a data model with a planning process.

1374
00:53:36,000 --> 00:53:38,000
You can have clean routes, sensible calendars,

1375
00:53:38,000 --> 00:53:40,480
and current order status, then still create chaos

1376
00:53:40,480 --> 00:53:42,880
if anyone can change priorities without agreement,

1377
00:53:42,880 --> 00:53:46,320
or if the floor receives updates through five different channels.

1378
00:53:46,320 --> 00:53:47,920
Before software automates decisions,

1379
00:53:47,920 --> 00:53:51,360
the organization needs to decide who owns those decisions

1380
00:53:51,360 --> 00:53:54,640
and how the plan changes when real life intervenes.

1381
00:53:54,640 --> 00:53:56,880
Fix the planning process before automating it.

1382
00:53:56,880 --> 00:53:59,600
Here's the problem most manufacturers don't talk about.

1383
00:53:59,600 --> 00:54:01,200
Before you even look at software,

1384
00:54:01,200 --> 00:54:03,680
you need to map how planning actually works today.

1385
00:54:03,680 --> 00:54:07,040
Not the official procedure or the ERP workflow,

1386
00:54:07,040 --> 00:54:08,640
I mean, who really changes priorities

1387
00:54:08,640 --> 00:54:10,240
on a difficult Tuesday afternoon,

1388
00:54:10,240 --> 00:54:11,440
who approves that change,

1389
00:54:11,440 --> 00:54:14,080
and how the person on the floor finds out about it.

1390
00:54:14,080 --> 00:54:17,120
Most factories discover that planning doesn't belong to one role.

1391
00:54:17,120 --> 00:54:19,680
Sales changes a delivery date after a customer call,

1392
00:54:19,680 --> 00:54:21,600
purchasing reports a delayed component,

1393
00:54:21,600 --> 00:54:23,120
production asks for a different sequence

1394
00:54:23,120 --> 00:54:24,720
because the machine needs attention,

1395
00:54:24,720 --> 00:54:26,400
quality places a batch on hold,

1396
00:54:26,400 --> 00:54:27,920
and the planner receives all of that

1397
00:54:27,920 --> 00:54:30,240
and tries to turn it into one workable plan.

1398
00:54:30,240 --> 00:54:31,520
Nobody's doing anything wrong there,

1399
00:54:31,520 --> 00:54:33,120
they're responding to real pressure,

1400
00:54:33,120 --> 00:54:35,840
but if every role can alter the plan through a different channel,

1401
00:54:35,840 --> 00:54:37,840
the factory doesn't have one planning process.

1402
00:54:37,840 --> 00:54:39,920
It has a series of well-ment interruptions.

1403
00:54:39,920 --> 00:54:42,240
Think of it like this, picture a normal morning.

1404
00:54:42,240 --> 00:54:44,960
Sales has promised an earlier delivery for a customer,

1405
00:54:44,960 --> 00:54:46,640
purchasing knows a supplier shipment

1406
00:54:46,640 --> 00:54:48,160
won't arrive until tomorrow.

1407
00:54:48,160 --> 00:54:50,160
A supervisor wants to keep a machine loaded

1408
00:54:50,160 --> 00:54:52,240
because the next setup takes time.

1409
00:54:52,240 --> 00:54:54,080
Meanwhile, the planner is staring at a late order

1410
00:54:54,080 --> 00:54:55,520
that needs the same resource,

1411
00:54:55,520 --> 00:54:58,160
and quality has just held part of its material for review.

1412
00:54:58,160 --> 00:54:59,840
Each person sees part of the problem,

1413
00:54:59,840 --> 00:55:01,200
and without shared rules,

1414
00:55:01,200 --> 00:55:02,720
each one may act on their own piece.

1415
00:55:02,720 --> 00:55:05,040
Sales changes a date,

1416
00:55:05,040 --> 00:55:06,480
purchasing sends an email,

1417
00:55:06,480 --> 00:55:09,040
the supervisor rearranges work on a local board,

1418
00:55:09,040 --> 00:55:10,960
and the planner updates a spreadsheet.

1419
00:55:10,960 --> 00:55:13,360
By lunchtime, people have four versions of the same day,

1420
00:55:13,360 --> 00:55:15,680
and none of them explains who approved the trade-offs.

1421
00:55:15,680 --> 00:55:18,320
That's a process problem before it becomes a software problem.

1422
00:55:18,320 --> 00:55:20,800
Start by asking who owns the plan at each level.

1423
00:55:20,800 --> 00:55:23,200
Someone needs responsibility for the broad production plan

1424
00:55:23,200 --> 00:55:24,480
across the coming weeks.

1425
00:55:24,480 --> 00:55:26,960
Someone needs responsibility for the detailed sequence

1426
00:55:26,960 --> 00:55:28,240
close to execution,

1427
00:55:28,240 --> 00:55:30,400
and the factory needs a clear path for exceptions

1428
00:55:30,400 --> 00:55:32,160
that affect customer commitments,

1429
00:55:32,160 --> 00:55:34,560
safety, quality, or constrained capacity.

1430
00:55:34,560 --> 00:55:37,520
Ownership doesn't mean one planner controls every decision alone.

1431
00:55:37,520 --> 00:55:39,760
It means people know where a decision belongs.

1432
00:55:39,760 --> 00:55:42,160
A supervisor can respond to a short local disruption

1433
00:55:42,160 --> 00:55:43,920
without calling a committee meeting.

1434
00:55:43,920 --> 00:55:46,640
A planner can revise the schedule when conditions change.

1435
00:55:46,640 --> 00:55:48,400
Sales can request priority,

1436
00:55:48,400 --> 00:55:50,480
but that request doesn't become a shop-flow instruction

1437
00:55:50,480 --> 00:55:52,880
until someone tests and approves the consequence.

1438
00:55:52,880 --> 00:55:54,720
Now you're ready for a planning cadence.

1439
00:55:54,720 --> 00:55:58,240
For many manufacturers, a rough plan works best on a weekly cycle.

1440
00:55:58,240 --> 00:55:59,840
It looks ahead at demand,

1441
00:55:59,840 --> 00:56:03,120
capacity pressure, known shortages, and major commitments.

1442
00:56:03,120 --> 00:56:05,200
The detailed plan may update daily,

1443
00:56:05,200 --> 00:56:07,840
because that's where sequences, staffing, work release,

1444
00:56:07,840 --> 00:56:10,240
and current resource status start to matter.

1445
00:56:10,240 --> 00:56:11,920
Then there are event-driven changes.

1446
00:56:11,920 --> 00:56:14,640
A machine breakdown, material shortage, quality hold,

1447
00:56:14,640 --> 00:56:16,320
urgent order, or staffing gap,

1448
00:56:16,320 --> 00:56:19,120
may require the plan to change outside the normal cycle.

1449
00:56:19,120 --> 00:56:20,080
That's fine.

1450
00:56:20,080 --> 00:56:22,080
Factories don't run on calendars alone.

1451
00:56:22,080 --> 00:56:23,920
But each event needs a rule.

1452
00:56:23,920 --> 00:56:26,000
Who reviews it, what information they need,

1453
00:56:26,000 --> 00:56:27,440
who can approve a change,

1454
00:56:27,440 --> 00:56:30,000
and how the revised priority reaches production.

1455
00:56:30,000 --> 00:56:33,280
Otherwise, every disruption turns into a fresh negotiation.

1456
00:56:33,280 --> 00:56:34,560
Here's the real challenge.

1457
00:56:34,560 --> 00:56:36,560
Commercial priority and physical feasibility

1458
00:56:36,560 --> 00:56:37,680
are different things.

1459
00:56:37,680 --> 00:56:39,600
A customer may matter deeply to the business

1460
00:56:39,600 --> 00:56:41,360
and they're automated serve priority,

1461
00:56:41,360 --> 00:56:43,280
but priority doesn't create machine time,

1462
00:56:43,280 --> 00:56:44,480
release held material,

1463
00:56:44,480 --> 00:56:46,400
or put a trained operator on the night shift.

1464
00:56:46,400 --> 00:56:48,160
The planning process needs to translate

1465
00:56:48,160 --> 00:56:50,480
the commercial request into a physical question.

1466
00:56:50,480 --> 00:56:52,240
What can move, what will slip,

1467
00:56:52,240 --> 00:56:53,920
and who accepts that result.

1468
00:56:53,920 --> 00:56:55,120
People often avoid this,

1469
00:56:55,120 --> 00:56:57,360
because it creates uncomfortable conversations.

1470
00:56:57,360 --> 00:56:59,520
Sales may not want to hear that an urgent order

1471
00:56:59,520 --> 00:57:01,680
pushes another customer beyond their date.

1472
00:57:01,680 --> 00:57:03,760
Production may not want to accept a sequence

1473
00:57:03,760 --> 00:57:05,200
that adds setup time.

1474
00:57:05,200 --> 00:57:07,440
Purchasing may not control the supplier delay,

1475
00:57:07,440 --> 00:57:10,400
yet hiding those consequences doesn't remove them.

1476
00:57:10,400 --> 00:57:13,520
It just lets the floor absorb them without a clear decision.

1477
00:57:13,520 --> 00:57:16,480
A workable process gives each exception a root.

1478
00:57:16,480 --> 00:57:19,200
For a shortage, the team needs to know whether to wait,

1479
00:57:19,200 --> 00:57:22,240
substitute, split the order, or change the sequence.

1480
00:57:22,240 --> 00:57:25,280
For a breakdown, they need to know whether an alternate resource exists,

1481
00:57:25,280 --> 00:57:26,640
whether overtime is possible,

1482
00:57:26,640 --> 00:57:28,160
or whether dates need revision.

1483
00:57:28,160 --> 00:57:29,600
For an urgent order, they need to know

1484
00:57:29,600 --> 00:57:31,600
who can approve the displacement of other work.

1485
00:57:31,600 --> 00:57:32,960
Keep those rules practical.

1486
00:57:32,960 --> 00:57:34,560
You don't need a thick governance document

1487
00:57:34,560 --> 00:57:36,160
that nobody opens again.

1488
00:57:36,160 --> 00:57:37,680
You need a shared habit.

1489
00:57:37,680 --> 00:57:39,760
A change enters through a known path,

1490
00:57:39,760 --> 00:57:41,280
the planner checks the impact,

1491
00:57:41,280 --> 00:57:42,800
the right person decides,

1492
00:57:42,800 --> 00:57:45,200
and production receives one approved instruction.

1493
00:57:45,200 --> 00:57:46,640
That's what software should support it.

1494
00:57:46,640 --> 00:57:49,120
If the current process relies on side conversations

1495
00:57:49,120 --> 00:57:50,560
and informal authority,

1496
00:57:50,560 --> 00:57:52,400
automating it only moves the confusion

1497
00:57:52,400 --> 00:57:54,320
into a more expensive place.

1498
00:57:54,320 --> 00:57:56,720
A planning system can calculate options quickly,

1499
00:57:56,720 --> 00:57:58,720
but it can't decide who has the right

1500
00:57:58,720 --> 00:58:01,760
to promise capacity that already belongs to another order.

1501
00:58:01,760 --> 00:58:03,600
So before comparing planning tools,

1502
00:58:03,600 --> 00:58:06,640
decide which planning decision needs to improve first.

1503
00:58:06,640 --> 00:58:09,040
Start with one decision, not a platform purchase.

1504
00:58:09,040 --> 00:58:10,880
Once you've mapped how the plan changes,

1505
00:58:10,880 --> 00:58:13,360
resist the urge to start with a platform shortlist.

1506
00:58:13,360 --> 00:58:14,480
I know that sounds backward,

1507
00:58:14,480 --> 00:58:16,960
a most planning projects begin with product demos,

1508
00:58:16,960 --> 00:58:19,360
but a platform purchase is a poor starting point

1509
00:58:19,360 --> 00:58:22,000
if nobody can name the decision the software should improve.

1510
00:58:22,000 --> 00:58:24,240
You'll get a lot of impressive scheduling screens,

1511
00:58:24,240 --> 00:58:26,320
and each one will seem useful until someone asks

1512
00:58:26,320 --> 00:58:28,320
what problem the plant is actually trying to solve

1513
00:58:28,320 --> 00:58:29,360
on Monday morning.

1514
00:58:29,360 --> 00:58:31,760
Start with one decision that keeps failing.

1515
00:58:31,760 --> 00:58:33,280
Maybe sales needs a credible answer

1516
00:58:33,280 --> 00:58:34,960
before promising a delivery date.

1517
00:58:34,960 --> 00:58:36,720
Maybe the planner needs to sequence work

1518
00:58:36,720 --> 00:58:38,800
each day on one constrained resource.

1519
00:58:38,800 --> 00:58:41,040
Maybe production releases orders too early,

1520
00:58:41,040 --> 00:58:43,040
only to discover that a material shortage

1521
00:58:43,040 --> 00:58:44,640
stops them halfway through the route.

1522
00:58:44,640 --> 00:58:47,280
Or perhaps several orders compete for one bottleneck,

1523
00:58:47,280 --> 00:58:49,360
and people need a consistent way to decide

1524
00:58:49,360 --> 00:58:51,280
which order receives the available time.

1525
00:58:51,280 --> 00:58:52,400
Those are different problems,

1526
00:58:52,400 --> 00:58:53,680
and they need different inputs,

1527
00:58:53,680 --> 00:58:54,320
different rules,

1528
00:58:54,320 --> 00:58:56,560
and sometimes different types of planning logic.

1529
00:58:56,560 --> 00:58:58,800
A promise date check needs demand,

1530
00:58:58,800 --> 00:59:01,280
route duration, resource availability,

1531
00:59:01,280 --> 00:59:02,880
and a view of current load.

1532
00:59:02,880 --> 00:59:04,640
Daily sequencing needs much more detail

1533
00:59:04,640 --> 00:59:06,160
about work already in progress,

1534
00:59:06,160 --> 00:59:07,280
setup conditions,

1535
00:59:07,280 --> 00:59:09,520
shift calendars in the next few operations.

1536
00:59:09,520 --> 00:59:11,760
Material constraint release depends on supply status

1537
00:59:11,760 --> 00:59:14,160
and the risk of creating work that will only wait on the floor.

1538
00:59:14,160 --> 00:59:15,840
So name the decision in plain language.

1539
00:59:15,840 --> 00:59:17,760
For example, when two late orders

1540
00:59:17,760 --> 00:59:19,520
need the same constrained machine,

1541
00:59:19,520 --> 00:59:20,720
we need to choose a sequence

1542
00:59:20,720 --> 00:59:23,120
that shows the delivery impact for both orders

1543
00:59:23,120 --> 00:59:24,720
before production starts.

1544
00:59:24,720 --> 00:59:26,160
That sentence does useful work.

1545
00:59:26,160 --> 00:59:27,600
It identifies the decision,

1546
00:59:27,600 --> 00:59:29,200
the condition that creates pressure

1547
00:59:29,200 --> 00:59:31,200
and the outcome the team needs to see.

1548
00:59:31,200 --> 00:59:32,800
It also stops the project from drifting

1549
00:59:32,800 --> 00:59:35,200
into a vague ambition to improve planning,

1550
00:59:35,200 --> 00:59:36,720
which usually means every department

1551
00:59:36,720 --> 00:59:38,960
brings a different expectation into the same meeting.

1552
00:59:38,960 --> 00:59:41,600
Now define the decision around five practical questions.

1553
00:59:41,600 --> 00:59:43,360
What information enters the decision?

1554
00:59:43,360 --> 00:59:45,600
That might include open order quantities,

1555
00:59:45,600 --> 00:59:46,640
customer dates,

1556
00:59:46,640 --> 00:59:48,160
current operation status,

1557
00:59:48,160 --> 00:59:49,360
machine availability,

1558
00:59:49,360 --> 00:59:50,240
setup family,

1559
00:59:50,240 --> 00:59:51,280
and material release,

1560
00:59:51,280 --> 00:59:52,960
which constraints can't be broken.

1561
00:59:52,960 --> 00:59:55,440
Perhaps one machine must perform the operation.

1562
00:59:55,440 --> 00:59:57,600
A tool is only available after a certain time

1563
00:59:57,600 --> 00:59:58,880
or a batch can't split.

1564
00:59:58,880 --> 01:00:00,240
What options are acceptable?

1565
01:00:00,240 --> 01:00:01,760
The team may permit overtime

1566
01:00:01,760 --> 01:00:03,200
and approved alternate route

1567
01:00:03,200 --> 01:00:05,360
or a later date for one class of orders.

1568
01:00:05,360 --> 01:00:07,200
Then ask who approves the choice.

1569
01:00:07,200 --> 01:00:09,120
The planning system can calculate options

1570
01:00:09,120 --> 01:00:11,920
but someone needs authority to accept the trade off.

1571
01:00:11,920 --> 01:00:14,080
If one sequence protects customer A

1572
01:00:14,080 --> 01:00:15,440
but delays customer B,

1573
01:00:15,440 --> 01:00:16,800
the answer may need a planner,

1574
01:00:16,800 --> 01:00:17,680
a production lead,

1575
01:00:17,680 --> 01:00:19,040
or a commercial owner,

1576
01:00:19,040 --> 01:00:20,160
depending on the rule.

1577
01:00:20,160 --> 01:00:21,680
Put that authority into the process

1578
01:00:21,680 --> 01:00:24,800
before the first automated recommendation arrives.

1579
01:00:24,800 --> 01:00:26,560
The last question is easy to miss.

1580
01:00:26,560 --> 01:00:28,000
What execution feedback tells you

1581
01:00:28,000 --> 01:00:29,360
whether the decision worked?

1582
01:00:29,360 --> 01:00:31,120
If the system recommends a sequence

1583
01:00:31,120 --> 01:00:32,480
and production follows it,

1584
01:00:32,480 --> 01:00:33,920
you need to know whether the job started

1585
01:00:33,920 --> 01:00:34,880
when expected,

1586
01:00:34,880 --> 01:00:36,480
whether the setup assumption held

1587
01:00:36,480 --> 01:00:37,840
and whether the planned completion date

1588
01:00:37,840 --> 01:00:39,840
survived contact with the shop floor.

1589
01:00:39,840 --> 01:00:42,160
Without that feedback, the model never improves.

1590
01:00:42,160 --> 01:00:44,640
It just repeats the same confident mistake faster,

1591
01:00:44,640 --> 01:00:46,320
which leads to a practical scenario.

1592
01:00:46,320 --> 01:00:48,880
Picture two orders waiting for the same machining center.

1593
01:00:48,880 --> 01:00:49,680
Both are late,

1594
01:00:49,680 --> 01:00:50,800
one belongs to a customer

1595
01:00:50,800 --> 01:00:52,480
with a delivery commitment tomorrow.

1596
01:00:52,480 --> 01:00:53,920
The other feeds a larger assembly

1597
01:00:53,920 --> 01:00:55,120
that ships later in the week

1598
01:00:55,120 --> 01:00:56,480
but only if machining finishes

1599
01:00:56,480 --> 01:00:59,120
before a downstream test slot becomes available.

1600
01:00:59,120 --> 01:01:01,760
A simple due date rule may push the first order ahead

1601
01:01:01,760 --> 01:01:03,840
but a planner may suspect that the second order

1602
01:01:03,840 --> 01:01:05,920
creates a bigger delivery risk if it waits.

1603
01:01:05,920 --> 01:01:08,640
This is a good first use case because the conflict is real.

1604
01:01:08,640 --> 01:01:11,360
The planning tool doesn't need to run the whole plant on day one.

1605
01:01:11,360 --> 01:01:13,200
It needs to compare the feasible sequences

1606
01:01:13,200 --> 01:01:14,320
for those two orders,

1607
01:01:14,320 --> 01:01:15,760
account for the actual machine time

1608
01:01:15,760 --> 01:01:17,120
in routing consequences

1609
01:01:17,120 --> 01:01:19,600
and show why one choice creates less harm than the other.

1610
01:01:19,600 --> 01:01:22,560
Then the planner can challenge the assumptions,

1611
01:01:22,560 --> 01:01:23,760
approve the result,

1612
01:01:23,760 --> 01:01:26,080
and explain the decision to production in sales.

1613
01:01:26,080 --> 01:01:27,840
That explanation matters more than people think

1614
01:01:27,840 --> 01:01:30,480
if the recommendation changes from one day to the next.

1615
01:01:30,480 --> 01:01:32,800
The planner should be able to explain what changed.

1616
01:01:32,800 --> 01:01:34,320
Maybe a machine lost capacity,

1617
01:01:34,320 --> 01:01:36,000
maybe material cleared quality,

1618
01:01:36,000 --> 01:01:38,960
maybe another order entered with a higher approved priority.

1619
01:01:38,960 --> 01:01:40,800
If the system can't explain the change in terms,

1620
01:01:40,800 --> 01:01:43,040
people recognize it won't become part of daily work.

1621
01:01:43,040 --> 01:01:45,840
A planning project should pass a simple test.

1622
01:01:45,840 --> 01:01:48,400
Can the planner describe the recommended decision?

1623
01:01:48,400 --> 01:01:49,680
The constraints behind it

1624
01:01:49,680 --> 01:01:51,760
and the consequences of choosing another option?

1625
01:01:51,760 --> 01:01:54,720
If not, the scope is probably too broad,

1626
01:01:54,720 --> 01:01:55,920
the data is too weak

1627
01:01:55,920 --> 01:01:58,880
or the software is solving a question nobody actually owns.

1628
01:01:58,880 --> 01:02:00,480
Once you start with one decision,

1629
01:02:00,480 --> 01:02:02,560
the next distinction becomes much clearer.

1630
01:02:02,560 --> 01:02:05,600
Planning, scheduling, optimization and simulation sound similar,

1631
01:02:05,600 --> 01:02:07,600
but they answer different factory questions.

1632
01:02:07,600 --> 01:02:09,120
Stick with me for the next section

1633
01:02:09,120 --> 01:02:11,760
because that distinction is where most projects get off track.

1634
01:02:11,760 --> 01:02:13,440
Planning, scheduling,

1635
01:02:13,440 --> 01:02:16,240
optimization and simulation are different jobs.

1636
01:02:16,240 --> 01:02:17,600
Let's cut through the hype.

1637
01:02:17,600 --> 01:02:19,040
Planning, scheduling,

1638
01:02:19,040 --> 01:02:21,200
optimization and simulation get thrown around

1639
01:02:21,200 --> 01:02:23,440
like they're all the same thing, but they really aren't.

1640
01:02:23,440 --> 01:02:25,760
That's where a lot of software discussions go off the rails.

1641
01:02:25,760 --> 01:02:28,320
A plant asks for better planning

1642
01:02:28,320 --> 01:02:31,200
and everyone in the room has a completely different job in mind.

1643
01:02:31,200 --> 01:02:32,720
One person wants an earlier warning

1644
01:02:32,720 --> 01:02:34,480
that demand exceeds capacity,

1645
01:02:34,480 --> 01:02:36,880
another wants a clean list of what to run next,

1646
01:02:36,880 --> 01:02:39,600
and a third wants help balancing customer service,

1647
01:02:39,600 --> 01:02:41,520
set up time and overtime.

1648
01:02:41,520 --> 01:02:42,880
Those needs overlap, sure,

1649
01:02:42,880 --> 01:02:45,120
but each needs its own kind of logic.

1650
01:02:45,120 --> 01:02:47,040
Production planning starts further out,

1651
01:02:47,040 --> 01:02:49,200
looking at what demand and available capacity

1652
01:02:49,200 --> 01:02:50,880
imply over a window of time,

1653
01:02:50,880 --> 01:02:52,960
maybe a few weeks, a month or longer,

1654
01:02:52,960 --> 01:02:55,280
depending on how the plant buys material,

1655
01:02:55,280 --> 01:02:57,440
staff's shifts and commits to customers?

1656
01:02:58,160 --> 01:03:00,960
Planning covers broad questions like whether there's enough capacity

1657
01:03:00,960 --> 01:03:02,000
for the order book,

1658
01:03:02,000 --> 01:03:03,520
where demand will pile up,

1659
01:03:03,520 --> 01:03:05,360
which work centers feel pressure,

1660
01:03:05,360 --> 01:03:06,560
when to release work,

1661
01:03:06,560 --> 01:03:08,960
and whether a new customer order creates a problem

1662
01:03:08,960 --> 01:03:11,360
that sales needs to discuss before promising a date.

1663
01:03:11,360 --> 01:03:13,680
It works with an aggregated view.

1664
01:03:13,680 --> 01:03:15,760
You might plan an hours per work center,

1665
01:03:15,760 --> 01:03:16,960
days per production area,

1666
01:03:16,960 --> 01:03:18,800
or capacity by product family.

1667
01:03:18,800 --> 01:03:20,640
That level of detail supports decisions

1668
01:03:20,640 --> 01:03:21,440
about overtime,

1669
01:03:21,440 --> 01:03:22,720
subcontracting, staffing,

1670
01:03:22,720 --> 01:03:24,720
purchasing and promised delivery dates,

1671
01:03:24,720 --> 01:03:26,160
but it doesn't always tell an operator

1672
01:03:26,160 --> 01:03:28,560
what job starts at 9 o'clock on Wednesday.

1673
01:03:28,560 --> 01:03:30,160
That next job belongs to scheduling.

1674
01:03:30,160 --> 01:03:32,000
Scheduling takes the broad production demand

1675
01:03:32,000 --> 01:03:34,640
and turns it into a more exact sequence of operations,

1676
01:03:34,640 --> 01:03:36,720
asking which order should use which resource,

1677
01:03:36,720 --> 01:03:38,800
in what order and at what time.

1678
01:03:38,800 --> 01:03:41,920
While respecting the limits the factory has decided to model.

1679
01:03:41,920 --> 01:03:43,360
Machine calendars,

1680
01:03:43,360 --> 01:03:44,480
shift patterns,

1681
01:03:44,480 --> 01:03:45,680
setup families,

1682
01:03:45,680 --> 01:03:46,400
tooling,

1683
01:03:46,400 --> 01:03:47,440
qualifications,

1684
01:03:47,440 --> 01:03:48,560
maintenance windows,

1685
01:03:48,560 --> 01:03:49,840
material readiness,

1686
01:03:49,840 --> 01:03:51,520
and the work already in progress.

1687
01:03:51,520 --> 01:03:53,360
A schedule sits closer to the floor,

1688
01:03:53,360 --> 01:03:54,800
dealing with the awkward reality

1689
01:03:54,800 --> 01:03:57,040
that two jobs may both need the same machine,

1690
01:03:57,040 --> 01:03:58,160
the same fixture,

1691
01:03:58,160 --> 01:04:00,320
or the same trained person at the same time.

1692
01:04:00,320 --> 01:04:01,920
A planning view might show

1693
01:04:01,920 --> 01:04:04,160
that the work center has enough hours across the week,

1694
01:04:04,160 --> 01:04:06,560
but a schedule must answer whether the jobs can run

1695
01:04:06,560 --> 01:04:08,960
in an order that actually fits inside those hours.

1696
01:04:08,960 --> 01:04:10,880
Think of planning as asking whether the factory

1697
01:04:10,880 --> 01:04:12,080
can carry the workload.

1698
01:04:12,080 --> 01:04:14,400
Scheduling asks how the factory will carry it.

1699
01:04:14,400 --> 01:04:16,480
Optimization enters the picture when the plant

1700
01:04:16,480 --> 01:04:18,160
has more than one feasible schedule

1701
01:04:18,160 --> 01:04:19,920
and needs help choosing between them.

1702
01:04:19,920 --> 01:04:20,800
It doesn't just ask,

1703
01:04:20,800 --> 01:04:22,240
can we run the sequence?

1704
01:04:22,240 --> 01:04:22,960
It asks,

1705
01:04:22,960 --> 01:04:26,080
which feasible sequence best supports the goals we have chosen?

1706
01:04:26,080 --> 01:04:27,440
Those goals need to be explicit.

1707
01:04:27,440 --> 01:04:29,200
You might want to reduce late orders,

1708
01:04:29,200 --> 01:04:31,120
cut changeovers on a constrained resource,

1709
01:04:31,120 --> 01:04:32,880
protect customer commitments,

1710
01:04:32,880 --> 01:04:34,080
reduce overtime,

1711
01:04:34,080 --> 01:04:35,440
or avoid releasing work

1712
01:04:35,440 --> 01:04:37,360
that will just sit between operations.

1713
01:04:37,360 --> 01:04:39,120
Sometimes those goals conflict.

1714
01:04:39,120 --> 01:04:40,800
A sequence that reduces setup time

1715
01:04:40,800 --> 01:04:42,320
may delay an urgent order,

1716
01:04:42,320 --> 01:04:44,080
but optimization makes those trade-offs

1717
01:04:44,080 --> 01:04:46,080
visible through rules and priorities.

1718
01:04:46,080 --> 01:04:48,080
It can search through many possible schedules

1719
01:04:48,080 --> 01:04:50,320
faster than a person can test them manually,

1720
01:04:50,320 --> 01:04:52,560
yet it doesn't discover your business priorities

1721
01:04:52,560 --> 01:04:53,520
by itself.

1722
01:04:53,520 --> 01:04:55,840
If you tell the engine that setup reduction matters

1723
01:04:55,840 --> 01:04:57,440
more than due date performance,

1724
01:04:57,440 --> 01:04:59,840
it may give you exactly that outcome.

1725
01:04:59,840 --> 01:05:01,440
Software can calculate the consequence

1726
01:05:01,440 --> 01:05:03,840
but people still need to choose what good means.

1727
01:05:03,840 --> 01:05:05,040
Then there's simulation.

1728
01:05:05,040 --> 01:05:06,960
Simulation does not create the life schedule

1729
01:05:06,960 --> 01:05:07,840
you intend to run.

1730
01:05:07,840 --> 01:05:09,680
Instead, it creates a controlled version

1731
01:05:09,680 --> 01:05:12,400
of a possible future based on assumptions you choose,

1732
01:05:12,400 --> 01:05:13,920
so the team can test the decision

1733
01:05:13,920 --> 01:05:16,080
before committing the real factory to it.

1734
01:05:16,080 --> 01:05:17,520
For example, imagine a plant expects

1735
01:05:17,520 --> 01:05:19,280
a major customer order next month,

1736
01:05:19,280 --> 01:05:21,520
but the order could arrive in several variants.

1737
01:05:21,520 --> 01:05:23,520
The planning team might simulate each variant

1738
01:05:23,520 --> 01:05:26,000
against current demand, expected material supply,

1739
01:05:26,000 --> 01:05:28,400
resource availability, and known maintenance work,

1740
01:05:28,400 --> 01:05:30,800
then see where each scenario creates risk.

1741
01:05:30,800 --> 01:05:32,480
Or consider a proposed second shift,

1742
01:05:32,480 --> 01:05:33,840
simulation can test what happens

1743
01:05:33,840 --> 01:05:35,280
if that shift adds machine time,

1744
01:05:35,280 --> 01:05:36,880
but not enough qualified labor,

1745
01:05:36,880 --> 01:05:39,120
or if more output from one area creates pressure

1746
01:05:39,120 --> 01:05:40,960
at a downstream inspection step.

1747
01:05:40,960 --> 01:05:42,640
It won't predict the future with certainty

1748
01:05:42,640 --> 01:05:44,320
because factories aren't that polite,

1749
01:05:44,320 --> 01:05:46,880
but it gives people a structured way to test assumptions

1750
01:05:46,880 --> 01:05:49,120
rather than arguing from instinct alone.

1751
01:05:49,120 --> 01:05:51,840
Let's use one factory question to make the difference clear.

1752
01:05:51,840 --> 01:05:54,000
A customer asks whether the plant can deliver new order

1753
01:05:54,000 --> 01:05:54,640
in three weeks.

1754
01:05:54,640 --> 01:05:56,800
Production planning asks whether the workload appears

1755
01:05:56,800 --> 01:05:59,440
to fit available capacity and material supply.

1756
01:05:59,440 --> 01:06:01,600
Scheduling asks whether new orders operations

1757
01:06:01,600 --> 01:06:04,080
would sit among jobs already competing for machines,

1758
01:06:04,080 --> 01:06:05,440
people, and tools.

1759
01:06:05,440 --> 01:06:08,240
Optimization asks which feasible sequence best protects

1760
01:06:08,240 --> 01:06:10,000
the dates and goals the business cares about,

1761
01:06:10,000 --> 01:06:11,920
including the impact on existing orders.

1762
01:06:11,920 --> 01:06:14,560
Simulation asks what happens if the supplier slips,

1763
01:06:14,560 --> 01:06:16,880
a resource loses time, demand changes,

1764
01:06:16,880 --> 01:06:18,960
or the plant adds temporary capacity.

1765
01:06:18,960 --> 01:06:20,400
Same order different questions.

1766
01:06:20,400 --> 01:06:22,160
This distinction matters when you evaluate

1767
01:06:22,160 --> 01:06:23,600
production planning software.

1768
01:06:23,600 --> 01:06:25,680
A tool built mainly for rough capacity planning

1769
01:06:25,680 --> 01:06:27,440
may give useful long-range warnings,

1770
01:06:27,440 --> 01:06:29,760
but struggle with detailed finite scheduling.

1771
01:06:29,760 --> 01:06:32,160
A detailed scheduler may create a workable sequence,

1772
01:06:32,160 --> 01:06:34,560
but not provide the scenario analysis management

1773
01:06:34,560 --> 01:06:36,880
once for a larger capacity decision.

1774
01:06:36,880 --> 01:06:40,080
And an optimization engine can produce a mathematically strong answer

1775
01:06:40,080 --> 01:06:42,480
that nobody uses if the objectives don't match

1776
01:06:42,480 --> 01:06:44,160
the way the plant actually runs.

1777
01:06:44,160 --> 01:06:46,080
So don't start by asking which product has

1778
01:06:46,080 --> 01:06:48,320
planning, scheduling, optimization, and simulation

1779
01:06:48,320 --> 01:06:49,520
written on its website.

1780
01:06:49,520 --> 01:06:52,400
Instead, ask which job your plant needs help with first,

1781
01:06:52,400 --> 01:06:54,720
because that choice has far more to do with variation

1782
01:06:54,720 --> 01:06:57,600
and constraints than with how many people work for the company.

1783
01:06:57,600 --> 01:07:00,240
Complexity, not headcount, drives the need.

1784
01:07:00,240 --> 01:07:02,640
Here's the real challenge most manufacturers face.

1785
01:07:02,640 --> 01:07:05,280
A small factory can lead serious planning support,

1786
01:07:05,280 --> 01:07:07,920
while a much larger factory may run fine with a board,

1787
01:07:07,920 --> 01:07:10,480
a spreadsheet, and people who know the work.

1788
01:07:10,480 --> 01:07:12,560
Company size doesn't settle the question.

1789
01:07:12,560 --> 01:07:15,120
I've seen the wrong buying trigger used again and again.

1790
01:07:15,120 --> 01:07:16,800
We've reached a certain number of employees,

1791
01:07:16,800 --> 01:07:19,120
so now we need an advanced planning system.

1792
01:07:19,120 --> 01:07:20,800
That may coincide with a real need,

1793
01:07:20,800 --> 01:07:24,080
but headcount tells you very little about the number of planning choices

1794
01:07:24,080 --> 01:07:27,120
people need to test before they release work.

1795
01:07:27,120 --> 01:07:28,960
Think about two very different plants.

1796
01:07:28,960 --> 01:07:31,760
The first runs are a repetitive product on a stable line,

1797
01:07:31,760 --> 01:07:33,280
where the root rarely changes,

1798
01:07:33,280 --> 01:07:35,360
and materials arrive on a predictable rhythm.

1799
01:07:35,360 --> 01:07:37,360
The same teams run the same assets,

1800
01:07:37,360 --> 01:07:40,320
and demand changes slowly enough that the supervisor can see

1801
01:07:40,320 --> 01:07:41,280
trouble coming.

1802
01:07:41,280 --> 01:07:42,720
That plant may employ many people,

1803
01:07:42,720 --> 01:07:44,960
but its planning problem stays fairly contained.

1804
01:07:44,960 --> 01:07:46,880
The second plant may employ far fewer people,

1805
01:07:46,880 --> 01:07:49,280
but builds to order, where each customer configuration

1806
01:07:49,280 --> 01:07:50,640
changes the root a little.

1807
01:07:50,640 --> 01:07:52,880
Several jobs share the same specialist machines, tools,

1808
01:07:52,880 --> 01:07:54,160
and inspection resources.

1809
01:07:54,160 --> 01:07:56,320
Some products need external processing,

1810
01:07:56,320 --> 01:07:58,240
and others can move through alternate paths,

1811
01:07:58,240 --> 01:08:00,160
but only under specific conditions.

1812
01:08:00,160 --> 01:08:02,800
A single, late component can change the feasible plan

1813
01:08:02,800 --> 01:08:04,160
for work already on the floor.

1814
01:08:04,160 --> 01:08:05,680
That is a different kind of workload.

1815
01:08:05,680 --> 01:08:08,640
The planning challenge grows when the number of dependencies grows,

1816
01:08:08,640 --> 01:08:10,640
and shared resources are one example.

1817
01:08:10,640 --> 01:08:13,440
If ten orders compete for one specialist operation,

1818
01:08:13,440 --> 01:08:16,160
the plan needs more than a broad capacity number.

1819
01:08:16,160 --> 01:08:17,760
They need to know the sequence effects,

1820
01:08:17,760 --> 01:08:19,920
the customer dates, the setup requirements,

1821
01:08:19,920 --> 01:08:22,240
and what each choice does to the rest of the root.

1822
01:08:22,240 --> 01:08:24,000
Rooting variation adds another layer.

1823
01:08:24,000 --> 01:08:26,160
A product family may look similar in ERP,

1824
01:08:26,160 --> 01:08:28,160
while the actual work differs by material,

1825
01:08:28,160 --> 01:08:29,840
revision, inspection level, coating,

1826
01:08:29,840 --> 01:08:31,680
packaging rule, or customer approval.

1827
01:08:31,680 --> 01:08:34,880
That doesn't mean the plant needs to turn every variation

1828
01:08:34,880 --> 01:08:36,240
into a software rule.

1829
01:08:36,240 --> 01:08:38,080
Instead, it means the plant needs to understand

1830
01:08:38,080 --> 01:08:40,320
which variations repeatedly change capacity

1831
01:08:40,320 --> 01:08:41,760
and sequencing decisions.

1832
01:08:41,760 --> 01:08:44,400
High mix and low volume often create this pressure early,

1833
01:08:44,400 --> 01:08:47,120
because each order can bring a different combination of constraints.

1834
01:08:47,120 --> 01:08:50,880
So the planer spends more time checking whether an old rule still applies.

1835
01:08:50,880 --> 01:08:53,440
By contrast, a high volume repetitive operation

1836
01:08:53,440 --> 01:08:54,880
may have a tighter production rhythm,

1837
01:08:54,880 --> 01:08:55,840
but fewer choices,

1838
01:08:55,840 --> 01:08:57,600
though it can still face planning problems,

1839
01:08:57,600 --> 01:09:00,640
especially around changeovers, supply, or bottlenecks.

1840
01:09:00,640 --> 01:09:04,080
The point is that volume alone doesn't explain the workload.

1841
01:09:04,080 --> 01:09:06,640
Bad rules can increase complexity as well.

1842
01:09:06,640 --> 01:09:08,960
Maybe work needs to stay together through heat treatment,

1843
01:09:08,960 --> 01:09:11,840
maybe it can split before assembly, but not after a quality check,

1844
01:09:11,840 --> 01:09:15,040
or a cleaning process might limit which materials can run together.

1845
01:09:15,040 --> 01:09:18,240
These conditions can turn a seemingly open capacity slot

1846
01:09:18,240 --> 01:09:20,960
into capacity that the order cannot actually use.

1847
01:09:20,960 --> 01:09:23,040
Subcontracting creates another planning boundary.

1848
01:09:23,040 --> 01:09:24,480
Once work leaves the plant,

1849
01:09:24,480 --> 01:09:26,480
the plan depends on outside lead times,

1850
01:09:26,480 --> 01:09:28,240
transport, supply or confirmation,

1851
01:09:28,240 --> 01:09:31,040
and the remaining operations waiting when the work returns.

1852
01:09:31,040 --> 01:09:33,840
A planer can manage that manually when the cases stay limited,

1853
01:09:33,840 --> 01:09:36,320
but when outside processing touches many orders.

1854
01:09:36,320 --> 01:09:39,360
Each supplier delay can create a chain of decisions inside the factory.

1855
01:09:39,360 --> 01:09:42,480
More sites and more shifts at coordination load,

1856
01:09:42,480 --> 01:09:45,040
even when the machines themselves don't change much.

1857
01:09:45,040 --> 01:09:47,920
A resource at another site may carry spare capacity,

1858
01:09:47,920 --> 01:09:50,400
but transfer time, material location, tooling,

1859
01:09:50,400 --> 01:09:54,240
local skills, and quality approval may limit whether that capacity helps.

1860
01:09:54,240 --> 01:09:56,560
Nightshift may create more available machine hours,

1861
01:09:56,560 --> 01:09:59,280
but less access to engineering or specialist support,

1862
01:09:59,280 --> 01:10:01,840
so the plan needs to reflect the operating conditions,

1863
01:10:01,840 --> 01:10:03,520
not just count hours.

1864
01:10:03,520 --> 01:10:05,920
Volatile demand has a similar effect.

1865
01:10:05,920 --> 01:10:09,520
If customers frequently change quantities, mix or requested dates,

1866
01:10:09,520 --> 01:10:13,360
the planning team must repeatedly check whether the revised demand still fits.

1867
01:10:13,360 --> 01:10:17,600
That becomes difficult when every change forces people to search through emails,

1868
01:10:17,600 --> 01:10:20,720
ERP notes, MESS status, supplier updates,

1869
01:10:20,720 --> 01:10:23,360
and local knowledge before they can even compare options.

1870
01:10:23,360 --> 01:10:24,960
That is a useful signal to watch.

1871
01:10:24,960 --> 01:10:28,960
When planers spend most of the day finding facts instead of weighing feasible choices,

1872
01:10:28,960 --> 01:10:31,040
the planning process has reached a limit.

1873
01:10:31,040 --> 01:10:32,880
The issue isn't that people lack skill,

1874
01:10:32,880 --> 01:10:37,280
but that the factory asks them to act as the connection layer between too many changing constraints.

1875
01:10:37,280 --> 01:10:40,640
Production planning software can reduce that search effort

1876
01:10:40,640 --> 01:10:43,440
if the model holds the relevant facts and relationships.

1877
01:10:43,440 --> 01:10:45,360
It can bring the choices into one place,

1878
01:10:45,360 --> 01:10:46,720
test them against agreed rules,

1879
01:10:46,720 --> 01:10:49,040
and show the planner where the real conflict sits.

1880
01:10:49,040 --> 01:10:52,080
It won't remove uncertainty and it won't make every change easy,

1881
01:10:52,080 --> 01:10:56,640
but it can stop people from rebuilding the planning picture from scratch every time conditions move.

1882
01:10:56,640 --> 01:11:00,240
Still, complexity by itself doesn't justify a software project.

1883
01:11:00,240 --> 01:11:02,160
A planning system brings work of its own,

1884
01:11:02,160 --> 01:11:06,800
data upkeep, rule ownership, integration, training, and ongoing support,

1885
01:11:06,800 --> 01:11:10,160
so the next question isn't only whether manual planning feels difficult,

1886
01:11:10,160 --> 01:11:14,800
it's whether the cost of keeping it manual now exceeds the cost and discipline of changing the process.

1887
01:11:14,800 --> 01:11:19,360
The cost of doing nothing and the cost of buying too early.

1888
01:11:19,360 --> 01:11:22,000
Here's the problem most manufacturers don't talk about.

1889
01:11:22,000 --> 01:11:23,520
Planning software costs money,

1890
01:11:23,520 --> 01:11:27,920
but so does keeping your current planning method when it can't keep up anymore.

1891
01:11:27,920 --> 01:11:29,600
The cost just shows up in a different place.

1892
01:11:29,600 --> 01:11:32,800
The cost of doing nothing doesn't show up on a single line in the budget.

1893
01:11:32,800 --> 01:11:35,520
It shows up in late orders, overtime approvals, rushed freight,

1894
01:11:35,520 --> 01:11:36,720
excess work in progress,

1895
01:11:36,720 --> 01:11:40,240
and people spending their whole day chasing updates instead of managing production.

1896
01:11:40,240 --> 01:11:41,840
When those things happen often enough,

1897
01:11:41,840 --> 01:11:43,440
they become the way the plant operates.

1898
01:11:43,440 --> 01:11:46,640
You start thinking of firefighting as normal because everybody's doing it.

1899
01:11:46,640 --> 01:11:47,920
Nobody questions it anymore.

1900
01:11:47,920 --> 01:11:50,240
Here's what a late order actually looks like on the ground.

1901
01:11:50,240 --> 01:11:53,040
Customer service calls sales, sales calls planning,

1902
01:11:53,040 --> 01:11:55,920
and planning asks production to interrupt the schedule,

1903
01:11:55,920 --> 01:11:58,400
which means production pulls people onto the urgent work,

1904
01:11:58,400 --> 01:12:00,400
and another order starts late right behind it.

1905
01:12:00,400 --> 01:12:03,920
The plant might recover one promise while quietly breaking another one.

1906
01:12:03,920 --> 01:12:06,160
That isn't always avoidable, I get it.

1907
01:12:06,160 --> 01:12:08,320
Factories deal with real disruption every day.

1908
01:12:08,320 --> 01:12:10,000
Equipment fails, suppliers slip,

1909
01:12:10,000 --> 01:12:11,600
and customers change their minds.

1910
01:12:11,600 --> 01:12:15,440
The real concern starts when the response becomes the same every single time.

1911
01:12:15,440 --> 01:12:18,160
Put the latest urgent order at the front, work longer,

1912
01:12:18,160 --> 01:12:19,760
and hope the rest sorts itself out.

1913
01:12:19,760 --> 01:12:21,920
That's not a strategy, it's a survival habit.

1914
01:12:21,920 --> 01:12:23,920
Work in progress often tells the same story.

1915
01:12:23,920 --> 01:12:25,280
When people don't trust the plan,

1916
01:12:25,280 --> 01:12:29,840
they release orders early just to keep everyone busy and material piles up between operations.

1917
01:12:29,840 --> 01:12:33,200
Jobs wait for machines, inspections, tools, or missing components.

1918
01:12:33,200 --> 01:12:36,400
The factory can look very busy while delivery performance stays unstable.

1919
01:12:36,400 --> 01:12:39,680
That's one of the most expensive forms of false comfort you'll ever see.

1920
01:12:39,680 --> 01:12:41,600
Planner workload carries a human cost too,

1921
01:12:41,600 --> 01:12:43,360
and it's bigger than most people realize.

1922
01:12:43,360 --> 01:12:46,400
A skilled planner can spend hours every day comparing dates,

1923
01:12:46,400 --> 01:12:48,960
calling supervisors, checking material status,

1924
01:12:48,960 --> 01:12:51,920
and updating files that go stale shortly after they save them.

1925
01:12:51,920 --> 01:12:53,840
Over time, planning becomes damage control.

1926
01:12:53,840 --> 01:12:57,200
The person with the most experience becomes the only one who can keep the work moving

1927
01:12:57,200 --> 01:13:00,640
and their capacity turns into a production constraint all by itself.

1928
01:13:00,640 --> 01:13:03,680
Yet buying software too early creates a completely different problem.

1929
01:13:03,680 --> 01:13:06,640
If the plant hasn't agreed on routing, actual resource rules,

1930
01:13:06,640 --> 01:13:09,280
setup assumptions, or authority for priority changes.

1931
01:13:09,280 --> 01:13:12,480
A new scheduling tool just formalizes the confusion.

1932
01:13:12,480 --> 01:13:14,080
The system may generate a schedule,

1933
01:13:14,080 --> 01:13:16,720
but people will override it because they know the rules inside it

1934
01:13:16,720 --> 01:13:18,800
don't reflect what actually happens on the floor.

1935
01:13:18,800 --> 01:13:21,360
That's not a software failure, that's a process gap,

1936
01:13:21,360 --> 01:13:23,440
and that's when the project gains a reputation.

1937
01:13:23,440 --> 01:13:25,440
Supervisors go back to their local boards,

1938
01:13:25,440 --> 01:13:27,440
planners keep the spreadsheet just for now,

1939
01:13:27,440 --> 01:13:29,200
which often means indefinitely.

1940
01:13:29,200 --> 01:13:31,120
Management sees a tool that nobody follows

1941
01:13:31,120 --> 01:13:33,760
and concludes that production planning software doesn't work,

1942
01:13:33,760 --> 01:13:38,400
when the real issue is that the project automated a model nobody had ever tested.

1943
01:13:38,400 --> 01:13:41,360
Consider a detailed scheduler introduced into a finishing area.

1944
01:13:41,360 --> 01:13:44,800
The tool receives order demand, operation times, and machine calendars,

1945
01:13:44,800 --> 01:13:46,880
and it creates an orderly sequence for the week.

1946
01:13:46,880 --> 01:13:50,320
But the model doesn't include actual clean-down time between material types,

1947
01:13:50,320 --> 01:13:53,680
because that rule lives with the team on the floor and never reach the project team.

1948
01:13:53,680 --> 01:13:57,440
Come Monday, production changes the schedule within the first few hours,

1949
01:13:57,440 --> 01:13:58,960
they aren't rejecting the software.

1950
01:13:58,960 --> 01:14:02,560
They're protecting the process from a plan that ignores a real condition.

1951
01:14:02,560 --> 01:14:06,160
If that happens repeatedly, users learn that the scheduler produces a starting point

1952
01:14:06,160 --> 01:14:08,080
at best and a distraction at worst.

1953
01:14:08,080 --> 01:14:11,520
The software becomes another system, people have to update without ever trusting.

1954
01:14:11,520 --> 01:14:14,000
The license cost is only one piece of the decision.

1955
01:14:14,000 --> 01:14:15,600
You also need time from planners,

1956
01:14:15,600 --> 01:14:18,320
supervisors, engineering, quality maintenance, and IT.

1957
01:14:18,320 --> 01:14:22,880
Someone has to clean up the first dataset and decide who owns planning rules after go live.

1958
01:14:22,880 --> 01:14:26,480
Integration needs design, testing, monitoring, and ongoing support.

1959
01:14:26,480 --> 01:14:27,520
People need training.

1960
01:14:27,520 --> 01:14:31,040
But more than that, they need time to test whether the recommended plan actually matches

1961
01:14:31,040 --> 01:14:32,560
how work really moves on the floor.

1962
01:14:32,560 --> 01:14:34,480
That's change cost, real change cost.

1963
01:14:34,480 --> 01:14:36,800
It's the cost of teaching people a new way to work,

1964
01:14:36,800 --> 01:14:39,360
and that's not something you can assign to a purchase order.

1965
01:14:39,360 --> 01:14:42,320
It's not something you can solve with a line item in a budget.

1966
01:14:42,320 --> 01:14:44,000
It's not an argument against software at all.

1967
01:14:44,000 --> 01:14:47,760
It's an argument against pretending that a purchase order solves a planning problem.

1968
01:14:47,760 --> 01:14:50,320
A planning system changes how people share information,

1969
01:14:50,320 --> 01:14:53,120
how they approve exceptions, and how they explain trade-offs.

1970
01:14:53,120 --> 01:14:56,160
If the current process can't name the decision, the constraints,

1971
01:14:56,160 --> 01:14:58,000
and the person who owns an exception,

1972
01:14:58,000 --> 01:15:00,320
software will only make the gaps easier to see.

1973
01:15:00,320 --> 01:15:01,840
So compare both costs honestly.

1974
01:15:01,840 --> 01:15:04,240
Doing nothing keeps charging the factory through late orders,

1975
01:15:04,240 --> 01:15:06,560
overtime, expediting, excess work and progress,

1976
01:15:06,560 --> 01:15:09,360
and planners who spend their day stitching together facts.

1977
01:15:09,360 --> 01:15:12,480
Buying two early charges the factory through a model people don't trust,

1978
01:15:12,480 --> 01:15:13,520
rules nobody owns,

1979
01:15:13,520 --> 01:15:15,840
and a second system that the floor works around.

1980
01:15:15,840 --> 01:15:18,640
The sensible point to act sits between those two extremes.

1981
01:15:18,640 --> 01:15:21,680
When manual planning creates repeatable operational damage,

1982
01:15:21,680 --> 01:15:24,160
and when the team can define one real decision,

1983
01:15:24,160 --> 01:15:25,440
the data it needs,

1984
01:15:25,440 --> 01:15:27,280
and the rules that make an answer feasible.

1985
01:15:27,280 --> 01:15:30,480
Then the first implementation has something concrete to prove.

1986
01:15:30,480 --> 01:15:32,880
A sensible implementation path.

1987
01:15:32,880 --> 01:15:34,320
So you've got the green light.

1988
01:15:34,320 --> 01:15:35,440
Here's a common trap.

1989
01:15:35,440 --> 01:15:38,080
Don't try to replace the whole planning process overnight.

1990
01:15:38,080 --> 01:15:40,480
Instead pick one area where the pain is obvious.

1991
01:15:40,480 --> 01:15:42,960
One bottleneck, one product family, one production cell,

1992
01:15:42,960 --> 01:15:45,200
or even just a single decision like whether a new order

1993
01:15:45,200 --> 01:15:46,720
can hit its requested date.

1994
01:15:46,720 --> 01:15:49,280
The point is to keep the first scope small

1995
01:15:49,280 --> 01:15:51,680
and tied to a real problem people are feeling right now.

1996
01:15:51,680 --> 01:15:54,480
Say you've got a machining group that's always the reason orders ship late

1997
01:15:54,480 --> 01:15:56,880
because multiple product families share the same equipment.

1998
01:15:56,880 --> 01:15:59,520
That's a solid starting point even if there's plenty else to fix.

1999
01:15:59,520 --> 01:16:00,800
You can model the demand,

2000
01:16:00,800 --> 01:16:02,720
routing's machine calendars, setup rules,

2001
01:16:02,720 --> 01:16:04,720
and actual exceptions around that group

2002
01:16:04,720 --> 01:16:07,680
without pretending you'll capture every factory rule on day one.

2003
01:16:07,680 --> 01:16:11,200
Before you touch any tool, document how planning works there now.

2004
01:16:11,200 --> 01:16:12,240
Talk to the planners.

2005
01:16:12,240 --> 01:16:13,600
How do they build the sequence?

2006
01:16:13,600 --> 01:16:17,040
Ask supervisors what they change after the plan hits the floor.

2007
01:16:17,040 --> 01:16:19,200
Then capture the exceptions that keep happening,

2008
01:16:19,200 --> 01:16:21,840
not just the dramatic one someone remembers from last month.

2009
01:16:21,840 --> 01:16:24,320
You need a baseline or you could spend months building a model

2010
01:16:24,320 --> 01:16:26,240
and never know if it actually reduced re-planning

2011
01:16:26,240 --> 01:16:28,960
or just moved the work from one spreadsheet to another.

2012
01:16:28,960 --> 01:16:30,720
It doesn't need a complicated scorecard,

2013
01:16:30,720 --> 01:16:33,120
just an honest record of what disrupts the plan,

2014
01:16:33,120 --> 01:16:36,000
how often decisions change and where people lose time.

2015
01:16:36,000 --> 01:16:37,680
Then build that first constrained model

2016
01:16:37,680 --> 01:16:38,880
with the people who know the work,

2017
01:16:38,880 --> 01:16:40,640
the planner sees the demand pressure.

2018
01:16:40,640 --> 01:16:43,280
The supervisor knows the resource's real capacity,

2019
01:16:43,280 --> 01:16:45,760
maintenance understands the downtime windows.

2020
01:16:45,760 --> 01:16:49,120
Engineering might know why a route has to follow a specific path

2021
01:16:49,120 --> 01:16:50,960
and quality knows the release conditions

2022
01:16:50,960 --> 01:16:52,320
that block the next operation.

2023
01:16:52,320 --> 01:16:54,400
Bring them into the rule designer.

2024
01:16:54,400 --> 01:16:56,960
It feels slower than just loading data into a tool

2025
01:16:56,960 --> 01:16:59,040
because it forces people to turn working knowledge

2026
01:16:59,040 --> 01:17:00,400
into explicit decisions.

2027
01:17:00,400 --> 01:17:02,560
But the discussion itself uncovers assumptions.

2028
01:17:02,560 --> 01:17:04,960
Maybe one person thinks two machines are interchangeable

2029
01:17:04,960 --> 01:17:07,760
until the supervisor points out one lacks a fixture.

2030
01:17:07,760 --> 01:17:10,000
Maybe an operation time in the routing looks fine

2031
01:17:10,000 --> 01:17:13,200
until production explains it excludes a recurring inspection weight.

2032
01:17:13,200 --> 01:17:15,520
The first model doesn't eat to settle every disagreement,

2033
01:17:15,520 --> 01:17:18,000
just enough agreed rules to create a plan,

2034
01:17:18,000 --> 01:17:19,840
people can challenge against real work.

2035
01:17:19,840 --> 01:17:21,360
Treat it as a working model,

2036
01:17:21,360 --> 01:17:23,280
not a final description of the plant.

2037
01:17:23,280 --> 01:17:25,120
When it produces a result that seems wrong,

2038
01:17:25,120 --> 01:17:26,880
don't instantly blame the software.

2039
01:17:26,880 --> 01:17:29,840
But don't accept it just because it came from a computer either.

2040
01:17:29,840 --> 01:17:32,080
Trace back the rule, the data and the assumption.

2041
01:17:32,080 --> 01:17:33,520
That's where trust gets built.

2042
01:17:33,520 --> 01:17:35,920
Run the new plan in parallel with the existing method

2043
01:17:35,920 --> 01:17:37,760
before you change shop floor control.

2044
01:17:37,760 --> 01:17:39,120
Let the current planning continue

2045
01:17:39,120 --> 01:17:42,080
while the new model calculates the same decision alongside it.

2046
01:17:42,080 --> 01:17:44,320
Then compare and focus on the differences.

2047
01:17:44,320 --> 01:17:46,240
Some differences reveal missing rules.

2048
01:17:46,240 --> 01:17:49,120
Others show the current method has habits nobody wrote down.

2049
01:17:49,120 --> 01:17:52,320
Occasionally the model exposes a conflict manual planning missed

2050
01:17:52,320 --> 01:17:54,800
because nobody had time to trace the full consequence.

2051
01:17:54,800 --> 01:17:57,280
All three outcomes help as long as the team reviews them

2052
01:17:57,280 --> 01:17:59,760
without turning every mismatch into a political argument.

2053
01:17:59,760 --> 01:18:02,160
Picture a plan are pulling together tomorrow's sequence

2054
01:18:02,160 --> 01:18:05,600
in the spreadsheet while the new system generates its own proposal.

2055
01:18:05,600 --> 01:18:07,200
The spreadsheet might put a rush order first

2056
01:18:07,200 --> 01:18:08,400
because sales call that morning.

2057
01:18:08,400 --> 01:18:10,080
The model might put a different order first

2058
01:18:10,080 --> 01:18:12,320
because delaying it would block a downstream assembly

2059
01:18:12,320 --> 01:18:13,600
that can't recover later.

2060
01:18:13,600 --> 01:18:15,440
Neither is automatically right.

2061
01:18:15,440 --> 01:18:18,800
The key is the conflict surfaces before machine time is committed

2062
01:18:18,800 --> 01:18:20,640
that gives the team a real conversation.

2063
01:18:20,640 --> 01:18:23,040
During the parallel phase capture overrides too.

2064
01:18:23,040 --> 01:18:26,560
If a planner or supervisor rejects the recommendation record why?

2065
01:18:26,560 --> 01:18:28,560
Maybe the system missed an operator limit

2066
01:18:28,560 --> 01:18:31,600
or an informal customer rule needs formal approval

2067
01:18:31,600 --> 01:18:34,080
or the person overriding can't explain it beyond

2068
01:18:34,080 --> 01:18:34,960
that's how we do it.

2069
01:18:34,960 --> 01:18:38,400
Those cases tell you what to model, what to govern

2070
01:18:38,400 --> 01:18:40,080
and what to leave to human judgment.

2071
01:18:40,080 --> 01:18:42,640
Don't move the new planning method into daily control

2072
01:18:42,640 --> 01:18:45,200
until planners trust the inputs understand the rules

2073
01:18:45,200 --> 01:18:46,800
and can explain the output to the floor.

2074
01:18:46,800 --> 01:18:50,160
Trust doesn't mean the plan never changes.

2075
01:18:50,160 --> 01:18:52,560
In a live factory, a plan that never changes

2076
01:18:52,560 --> 01:18:54,560
is probably ignoring the factory.

2077
01:18:54,560 --> 01:18:56,480
Trust means people know what changed and why.

2078
01:18:56,480 --> 01:18:58,880
Expand based on evidence, not just enthusiasm.

2079
01:18:58,880 --> 01:19:01,120
If the first area improves the targeted decision

2080
01:19:01,120 --> 01:19:02,720
and the team keeps the model current

2081
01:19:02,720 --> 01:19:04,240
then extend to the next constraint,

2082
01:19:04,240 --> 01:19:06,480
product group or related planning decision.

2083
01:19:06,480 --> 01:19:08,720
Each expansion adds relationships and rules

2084
01:19:08,720 --> 01:19:10,960
so give it the same care as the first scope.

2085
01:19:10,960 --> 01:19:13,200
This is not a big bang replacement of ERP,

2086
01:19:13,200 --> 01:19:14,720
MES or the planner's judgment.

2087
01:19:14,720 --> 01:19:17,120
It's a gradual way to put repeatable planning logic

2088
01:19:17,120 --> 01:19:20,000
where people can test, improve and use it under pressure.

2089
01:19:20,000 --> 01:19:22,320
Once that work starts producing dependable facts

2090
01:19:22,320 --> 01:19:25,040
you can also connect it to the wider Microsoft architecture

2091
01:19:25,040 --> 01:19:28,160
without asking a reporting platform to become a production scheduler.

2092
01:19:28,160 --> 01:19:31,280
Where Microsoft Fabric fits.

2093
01:19:31,280 --> 01:19:33,040
So once your planning model is trusted,

2094
01:19:33,040 --> 01:19:35,360
Microsoft Fabric can help tie that work

2095
01:19:35,360 --> 01:19:37,040
into the broader data estate.

2096
01:19:37,040 --> 01:19:39,920
Fabric wraps around the planning engine, not in place of it.

2097
01:19:39,920 --> 01:19:42,560
Consider all the sources feeding into a production decision.

2098
01:19:42,560 --> 01:19:46,080
ERP holds orders, routines, inventory and purchasing status.

2099
01:19:46,080 --> 01:19:48,080
MES has execution facts from the floor,

2100
01:19:48,080 --> 01:19:50,160
quality has release and holds status,

2101
01:19:50,160 --> 01:19:52,480
maintenance has planned downtime and asset events,

2102
01:19:52,480 --> 01:19:55,440
IoT data shows machine state, cycle signals, alarms,

2103
01:19:55,440 --> 01:19:57,280
each answers part of the question.

2104
01:19:57,280 --> 01:20:00,400
Fabric can pull those facts into a governed data foundation.

2105
01:20:00,400 --> 01:20:02,480
Now planners, data teams and operations leaders

2106
01:20:02,480 --> 01:20:05,280
all work from the same history instead of separate extracts.

2107
01:20:05,280 --> 01:20:07,920
That helps when the planning model needs a consistent feed

2108
01:20:07,920 --> 01:20:10,800
of current order status, run times, material risk

2109
01:20:10,800 --> 01:20:12,640
or capacity loss from a bottleneck.

2110
01:20:12,640 --> 01:20:15,520
But let's be clear, Fabric doesn't generate a feasible schedule.

2111
01:20:15,520 --> 01:20:17,680
It stores data, transforms it, governs access

2112
01:20:17,680 --> 01:20:18,960
and supports analytics.

2113
01:20:18,960 --> 01:20:21,040
It connects the dots between IT and OT.

2114
01:20:21,040 --> 01:20:22,800
Without a planning model and explicit rules,

2115
01:20:22,800 --> 01:20:26,960
though it can't decide if order 847 should run before order 848,

2116
01:20:26,960 --> 01:20:28,480
whether a setup sequence works

2117
01:20:28,480 --> 01:20:31,120
or if an alternate machine meets a customer requirement.

2118
01:20:31,120 --> 01:20:32,720
That logic stays in the planning engine

2119
01:20:32,720 --> 01:20:34,320
and in the rules your factory owns.

2120
01:20:34,320 --> 01:20:35,520
This distinction matters.

2121
01:20:35,520 --> 01:20:38,480
I see manufacturers assume that just pulling ERP,

2122
01:20:38,480 --> 01:20:42,080
MES and IoT data into one place automatically

2123
01:20:42,080 --> 01:20:43,680
creates decision intelligence.

2124
01:20:43,680 --> 01:20:44,400
It doesn't.

2125
01:20:44,400 --> 01:20:46,080
Share data is a better starting point,

2126
01:20:46,080 --> 01:20:47,360
but you still need a decision model

2127
01:20:47,360 --> 01:20:50,960
to turn that data into a tested choice under real constraints.

2128
01:20:50,960 --> 01:20:53,920
In practice, Fabric supports the planning process in three areas.

2129
01:20:53,920 --> 01:20:55,600
First, it creates a common data history.

2130
01:20:55,600 --> 01:20:57,280
Instead of fighting over which system

2131
01:20:57,280 --> 01:20:59,520
has the right answer when a late order appears,

2132
01:20:59,520 --> 01:21:02,160
the team can trace planned dates, actual starts,

2133
01:21:02,160 --> 01:21:04,000
actual finishes, material events,

2134
01:21:04,000 --> 01:21:06,800
quality holds an equipment downtime across systems.

2135
01:21:06,800 --> 01:21:09,440
That history helps improve planning assumptions.

2136
01:21:09,440 --> 01:21:12,480
Say your planning model expects an operation to start on time,

2137
01:21:12,480 --> 01:21:15,600
but it always runs late when a specific component hits the root.

2138
01:21:15,600 --> 01:21:17,840
Fabric can pull together order history,

2139
01:21:17,840 --> 01:21:21,200
material receipts, MES status and quality events

2140
01:21:21,200 --> 01:21:23,280
so the team can check if the issue is poor,

2141
01:21:23,280 --> 01:21:25,360
operation times, a release rule,

2142
01:21:25,360 --> 01:21:27,920
supplier uncertainty, or something else entirely.

2143
01:21:27,920 --> 01:21:31,280
That's far better than changing a date in a spreadsheet and forgetting why.

2144
01:21:31,840 --> 01:21:36,320
Second, Fabric supports data quality checks before the data reaches the scheduling process.

2145
01:21:36,320 --> 01:21:39,280
A root might reference a resource that no longer runs that work.

2146
01:21:39,280 --> 01:21:41,440
A production order might carry an old revision.

2147
01:21:41,440 --> 01:21:44,560
MES completions might not match the remaining quantity in ERP.

2148
01:21:44,560 --> 01:21:46,720
These aren't minor details when they feed a schedule,

2149
01:21:46,720 --> 01:21:48,080
they change the decision.

2150
01:21:48,080 --> 01:21:49,440
With a shared data layer,

2151
01:21:49,440 --> 01:21:50,800
teams can spot those issues

2152
01:21:50,800 --> 01:21:52,800
and root them to the right people.

2153
01:21:52,800 --> 01:21:54,800
The goal isn't a giant central data project

2154
01:21:54,800 --> 01:21:56,240
before planning can start.

2155
01:21:56,240 --> 01:21:58,240
It's to make recurring data issues visible,

2156
01:21:58,240 --> 01:22:00,000
traceable and easier to fix at the source.

2157
01:22:00,720 --> 01:22:04,240
Third, Fabric supports plan versus actual analysis over time.

2158
01:22:04,240 --> 01:22:07,680
The planning system produces an expected sequence and expected dates.

2159
01:22:07,680 --> 01:22:09,840
The factory executes under real conditions.

2160
01:22:09,840 --> 01:22:14,400
Fabric collects both views and helps the team investigate where the plan consistently drifts.

2161
01:22:14,400 --> 01:22:17,440
Maybe a bottleneck loses more time than the calendar accounts for.

2162
01:22:17,440 --> 01:22:19,200
Maybe a setup rule needs more detail.

2163
01:22:19,200 --> 01:22:21,360
Maybe orders release too early and then wait.

2164
01:22:21,360 --> 01:22:23,920
The planning engine needs those findings to improve its rules.

2165
01:22:23,920 --> 01:22:29,120
Operations leaders need them to see if the process itself creates recurring disruption.

2166
01:22:29,120 --> 01:22:31,120
So here's how the architecture comes together.

2167
01:22:31,120 --> 01:22:35,680
ERP stays the system of record for orders, materials and commercial commitments.

2168
01:22:35,680 --> 01:22:40,080
MES stays close to execution, traceability and shop floor status.

2169
01:22:40,080 --> 01:22:43,440
Maintenance, quality, engineering and IoT keep their own roles.

2170
01:22:43,440 --> 01:22:45,840
Fabric creates a shared data foundation across them.

2171
01:22:45,840 --> 01:22:47,600
Then the planning engine pulls the facts it needs

2172
01:22:47,600 --> 01:22:49,520
and applies the factory's constraint logic.

2173
01:22:49,520 --> 01:22:51,360
That separation keeps every piece honest.

2174
01:22:51,360 --> 01:22:54,560
Fabric gives governed data, history and analysis.

2175
01:22:54,560 --> 01:22:57,040
The planning system gives feasible scheduling decisions

2176
01:22:57,040 --> 01:23:01,040
and people still own the rules, the exceptions and the trade-offs when no option looks good.

2177
01:23:01,040 --> 01:23:06,480
As a Microsoft MVP, I spend a lot of time figuring out where fabric fits in industrial architecture.

2178
01:23:06,480 --> 01:23:10,800
For production planning, its strongest role is often before and after the scheduling decision.

2179
01:23:10,800 --> 01:23:15,200
Prepping trusted inputs, tracing changes and learning from actual execution.

2180
01:23:15,200 --> 01:23:17,920
But if every planning problem just ends in another report,

2181
01:23:17,920 --> 01:23:20,640
the factory still needs someone to decide what runs next.

2182
01:23:20,640 --> 01:23:24,080
Power BI, visibility before and before and after the decision.

2183
01:23:24,080 --> 01:23:27,200
Here's the thing about power BI in planning. It sits around the decision,

2184
01:23:27,200 --> 01:23:31,360
not inside the scheduling engine. That distinction clears up a lot of confusion from the start.

2185
01:23:31,360 --> 01:23:33,680
A report can flag an overloaded work center,

2186
01:23:33,680 --> 01:23:36,400
orders at risk or material shortages that keep stopping release,

2187
01:23:36,400 --> 01:23:41,040
but it won't test every feasible sequence across machines, tools, operators and setup rules.

2188
01:23:41,040 --> 01:23:44,960
Instead, it tells people where to look and whether the planning process is improving.

2189
01:23:44,960 --> 01:23:45,920
That's its job.

2190
01:23:45,920 --> 01:23:49,840
Before you lock in a plan, power BI can give planners and production leaders a shared view

2191
01:23:49,840 --> 01:23:51,840
of where the system is starting to buckle.

2192
01:23:51,840 --> 01:23:55,360
You might track backlog by work center, load against usable capacity,

2193
01:23:55,360 --> 01:24:00,080
orders nearing their due date, material shortages or maintenance windows that eat into constraint time.

2194
01:24:00,080 --> 01:24:02,880
The goal isn't another dashboard for the sake of having one.

2195
01:24:02,880 --> 01:24:07,120
The goal is to get everyone in the planning conversation starting from the same operating facts,

2196
01:24:07,120 --> 01:24:09,280
picture a daily review for a machining area.

2197
01:24:09,280 --> 01:24:11,280
The team can see several orders due soon,

2198
01:24:11,280 --> 01:24:13,520
but the real issue isn't total order count.

2199
01:24:13,520 --> 01:24:18,000
A small group of jobs depends on one operation that has less available time than the plan assumed.

2200
01:24:18,000 --> 01:24:21,120
Meanwhile, another group was released early and sits upstream,

2201
01:24:21,120 --> 01:24:24,480
adding queue time without helping delivery that changes the discussion.

2202
01:24:24,480 --> 01:24:29,360
Instead of asking why the area looks busy, the planner can ask which orders face a real delivery risk,

2203
01:24:29,360 --> 01:24:30,960
what constraint is causing it,

2204
01:24:30,960 --> 01:24:34,480
and whether releasing more work will help or just pile up work in progress.

2205
01:24:34,480 --> 01:24:38,400
Power BI can bring that context together from the planning system,

2206
01:24:38,400 --> 01:24:43,200
ERP, MES, quality and maintenance sources that gives people visibility before they act.

2207
01:24:43,200 --> 01:24:46,160
After the decision, power BI becomes just as useful.

2208
01:24:46,160 --> 01:24:48,240
A planning engine may propose a schedule,

2209
01:24:48,240 --> 01:24:53,440
and the team may approve it, but the factory still needs to learn whether the assumptions behind that schedule

2210
01:24:53,440 --> 01:24:55,040
held up during execution.

2211
01:24:55,040 --> 01:24:57,200
Compare planned and actual start times,

2212
01:24:57,200 --> 01:24:59,760
compare expected completion with actual completion,

2213
01:24:59,760 --> 01:25:03,440
check how often the work center ran a different sequence from the approved plan

2214
01:25:03,440 --> 01:25:05,600
and record why when the reason is known.

2215
01:25:05,600 --> 01:25:08,080
Those comparisons should improve the planning rules,

2216
01:25:08,080 --> 01:25:10,160
not punish people for normal disruption.

2217
01:25:10,160 --> 01:25:14,240
Suppose an operation keeps starting late even though the scheduler places it early enough in the day.

2218
01:25:14,240 --> 01:25:17,520
At first, someone may assume the resource calendar needs more buffer,

2219
01:25:17,520 --> 01:25:21,360
but when you connect the planned schedule with MES progress and material status,

2220
01:25:21,360 --> 01:25:22,880
a different pattern may show up.

2221
01:25:22,880 --> 01:25:26,800
The operation starts late because the order often waits for a prior inspection release.

2222
01:25:26,800 --> 01:25:29,840
That points to a release rule or a process dependency.

2223
01:25:29,840 --> 01:25:33,760
Changing the machine calendar would hide the symptom while the real problem continues.

2224
01:25:33,760 --> 01:25:36,560
This is where planned versus actual analysis gets practical.

2225
01:25:36,560 --> 01:25:39,840
It helps the team ask whether the model used the wrong duration,

2226
01:25:39,840 --> 01:25:42,400
missed a dependency, received status too late,

2227
01:25:42,400 --> 01:25:45,600
or planned work before the factory could physically release it.

2228
01:25:45,600 --> 01:25:49,600
A report can also surface patterns that stay invisible inside day-to-day firefighting.

2229
01:25:49,600 --> 01:25:54,400
Maybe planners override the proposed sequence most often after a certain type of quality hold.

2230
01:25:54,400 --> 01:25:57,280
Maybe the same product family misses its planned start,

2231
01:25:57,280 --> 01:25:59,840
whenever a particular tool is shared across lines.

2232
01:25:59,840 --> 01:26:03,920
Or maybe a work center shows decent OEE while order completion dates still drift

2233
01:26:03,920 --> 01:26:05,680
because the queue arrives in the wrong order.

2234
01:26:05,680 --> 01:26:06,880
Those are planning questions.

2235
01:26:06,880 --> 01:26:11,120
Power BI gives the team a way to see repeated exceptions across weeks and months.

2236
01:26:11,120 --> 01:26:13,920
Instead of treating every late order as an isolated event,

2237
01:26:13,920 --> 01:26:16,640
it can help production, planning, quality and maintenance,

2238
01:26:16,640 --> 01:26:19,600
discuss the same pattern with a shared history behind it, still.

2239
01:26:19,600 --> 01:26:22,160
Don't let a dashboard become the decision maker.

2240
01:26:22,160 --> 01:26:25,360
A red indicator doesn't tell an operator which job should run next.

2241
01:26:25,360 --> 01:26:28,720
A capacity chart doesn't determine whether the plan should accept overtime,

2242
01:26:28,720 --> 01:26:30,560
move work through an approved alternate route,

2243
01:26:30,560 --> 01:26:32,000
or renegotiate a customer date.

2244
01:26:32,000 --> 01:26:37,120
Those choices need planning rules, operational judgment, and clear authority.

2245
01:26:37,120 --> 01:26:40,080
Think of Power BI as the feedback loop around the decision.

2246
01:26:40,080 --> 01:26:43,680
It helps reveal where the plan faces pressure before execution.

2247
01:26:43,680 --> 01:26:46,960
It helps compare the plan result with the actual result afterward.

2248
01:26:46,960 --> 01:26:50,160
And it gives the factory evidence to adjust, routes, durations,

2249
01:26:50,160 --> 01:26:53,520
release rules or resource assumptions when the same failure keeps returning.

2250
01:26:53,520 --> 01:26:57,280
Once that visibility exists, the next question gets more practical.

2251
01:26:57,280 --> 01:26:59,120
How do you capture exceptions, route approvals,

2252
01:26:59,120 --> 01:27:01,760
and keep people working from the same revised plan without turning email

2253
01:27:01,760 --> 01:27:03,120
into the unofficial production system?

2254
01:27:03,120 --> 01:27:06,480
Power Platform, co-pilot and workflow boundaries.

2255
01:27:06,480 --> 01:27:09,200
Once you can see planning, pressure and plan adherence,

2256
01:27:09,200 --> 01:27:12,320
you still need a way to handle the exceptions that change the plan.

2257
01:27:12,320 --> 01:27:13,920
That's where Power Platform fits in.

2258
01:27:13,920 --> 01:27:15,680
Not as a replacement for a scheduling engine,

2259
01:27:15,680 --> 01:27:17,520
but as the working layer around it,

2260
01:27:17,520 --> 01:27:19,920
where people capture an issue, check the impact,

2261
01:27:19,920 --> 01:27:23,680
approve a decision, and make sure the right team receives the revised instruction.

2262
01:27:23,680 --> 01:27:25,760
Take a material shortage.

2263
01:27:25,760 --> 01:27:29,200
Someone in purchasing learns that a component won't arrive as expected.

2264
01:27:29,200 --> 01:27:32,640
Without a defined workflow, that information may travel through an email,

2265
01:27:32,640 --> 01:27:36,480
a team's message, a call to the planner, and maybe a note in ERP.

2266
01:27:36,480 --> 01:27:37,840
By the time the schedule changes,

2267
01:27:37,840 --> 01:27:40,800
the floor may have already released work that depends on the missing part.

2268
01:27:40,800 --> 01:27:43,760
A Power app can give that shortage a structured entry point.

2269
01:27:43,760 --> 01:27:46,000
The buyer can identify the affected material,

2270
01:27:46,000 --> 01:27:48,400
expected receipt date, affected orders,

2271
01:27:48,400 --> 01:27:49,840
and confidence in the new date.

2272
01:27:49,840 --> 01:27:53,600
The planner receives a planning exception instead of another unstructured message.

2273
01:27:53,600 --> 01:27:57,040
If the shortage affects a customer promise or a constrained resource,

2274
01:27:57,040 --> 01:27:59,600
the workflow can route it to the right person for review.

2275
01:27:59,600 --> 01:28:01,200
The app doesn't decide the schedule.

2276
01:28:01,200 --> 01:28:03,120
It captures the facts needed to revise it.

2277
01:28:03,120 --> 01:28:06,800
Power Automate can then help move that exception through an agreed process.

2278
01:28:06,800 --> 01:28:10,240
It can notify the planner, request approval for an overtime decision,

2279
01:28:10,240 --> 01:28:13,360
record who accepted a delivery impact or alert production

2280
01:28:13,360 --> 01:28:15,280
when an approved sequence changes.

2281
01:28:15,280 --> 01:28:17,840
That matters because a revised plan only helps

2282
01:28:17,840 --> 01:28:20,080
if people hear about it through a channel they trust.

2283
01:28:20,080 --> 01:28:22,800
Otherwise, the planning system may hold the latest version

2284
01:28:22,800 --> 01:28:24,800
while the floor runs yesterday's priorities.

2285
01:28:24,800 --> 01:28:26,480
This also applies to urgent orders.

2286
01:28:26,480 --> 01:28:30,800
Sales can submit a priority request with a requested date and customer reason.

2287
01:28:30,800 --> 01:28:33,760
The workflow can require someone to review the planning impact

2288
01:28:33,760 --> 01:28:36,080
before that request becomes a production instruction

2289
01:28:36,080 --> 01:28:38,640
that doesn't slow the business down for the sake of process.

2290
01:28:38,640 --> 01:28:41,600
It stops urgency from becoming an automatic overview.