Rough-Cut vs Detailed Capacity Planning

Rough-cut capacity planning checks aggregate demand against a few critical resources to see if a plan is roughly feasible, while detailed capacity planning schedules every operation finitely to produce an executable minute-by-minute plan. Rough-cut is fast and approximate and answers "can we probably do this." Detailed is precise and executable and answers "exactly when will each job start and finish, on which machine." They sit at different horizons and serve different decisions, and mature planning uses both: rough-cut to keep infeasible plans out of the shop, detailed to run the shop. EDGEBIC by User Solutions is a detailed finite engine, and it also supports the fast feasibility questions rough-cut is built for.

Two questions, two altitudes

Capacity planning is really two questions asked at two altitudes.

The high-altitude question is feasibility. Over the next month or quarter, will the total demand overwhelm the resources that actually limit you? You do not need every operation timed to answer that. You need the load on your two or three bottleneck work centers compared against their available hours in each bucket. That is rough-cut capacity planning: sum the hours, compare to the capacity, flag the overloads.

The low-altitude question is execution. Given a plan that is feasible in aggregate, what is the exact sequence each machine runs, when does each job start, and which jobs slip if capacity is tight? That requires detailed capacity planning: load every operation of every routing against finite work centers, honor setups and dependencies, and produce a plan operators can follow. This is the domain of finite capacity scheduling.

The core difference: buckets vs sequence

The sharpest way to see the difference is what each method does with time.

Rough-cut works in buckets. It pools demand into a week or a month, sums the required hours per resource, and compares that total to the resource's available hours for the bucket. If the bucket has enough hours, rough-cut calls it feasible. It does not care about the order jobs run in or whether the demand clusters at the start of the bucket.

Detailed works in sequence. It places each operation at a specific start time, so it knows that a work center with enough total hours in a month can still run jobs late if the demand lands in week one while the capacity is spread across four weeks. It accounts for setups, queue time, and the fact that a downstream step cannot start until its upstream step finishes.

That distinction is why the two can disagree, and why the detailed plan is the one that reflects reality on the floor.

A worked example

A heat-treat oven is your bottleneck. It runs 8 hours a day, 5 days a week, so 40 hours per week and about 160 hours in a 4-week month. This month's demand on the oven totals 150 hours.

Rough-cut view: 150 hours of load against 160 hours of capacity. That is 94 percent loaded, under capacity, so the month is feasible. Green light.

Detailed view: the demand is not spread evenly. Two big jobs, 45 hours combined, are both due in week one, and week one has only 40 oven hours available. The detailed finite schedule places the operations in sequence, discovers that week one is over capacity by 5 hours, and pushes part of the second job into week two. One order finishes 2 days late even though the month, in total, had 10 hours to spare.

Same oven, same 150 hours, two different answers. Rough-cut was right that the month fits. Detailed was right that the timing does not. You need the rough-cut read to know the month is not hopeless, and you need the detailed plan to see the week-one collision and fix it, by adding an oven shift, resequencing, or moving a date.

How the two fit together in practice

The best workflow chains them.

Start with rough-cut when the question is fast and strategic: should we accept this large order, does the master schedule stand a chance, which quote scenario is even plausible. A rough-cut check against your constraints answers in seconds and keeps clearly infeasible plans from ever reaching the detailed engine. This is close in spirit to the difference between finite and infinite capacity in practice: rough-cut often uses an infinite-loading style sum to get a quick read, then detailed applies the finite constraint.

Move to detailed when the plan has passed the feasibility gate and you need the executable sequence. The finite engine produces the actual Gantt, the exact completion dates, and the list of which existing jobs a new order displaces. For a customer promise, a quote simulation runs the same detailed logic on a hypothetical order and returns a realistic date without disturbing your live plan.

Why not just run detailed for everything

You can, and the detailed plan is always the source of truth. But rough-cut earns its place for speed and framing. When sales needs an answer while the customer is on the phone, a rough-cut read on the bottleneck is faster than a full reschedule and usually good enough to say "yes with confidence" or "that is going to be tight." When you are comparing five what-if scenarios, rough-cut screens them so you only run the detailed engine on the two that survive.

Think of rough-cut as the triage and detailed as the treatment. Triage tells you which problems are real and roughly how bad. Treatment gives you the exact plan. Use rough-cut to avoid wasting the detailed engine on hopeless plans, and use detailed to actually run the ones that matter. See the detailed finite engine work on your own routings in EDGEBIC, and read how load meets capacity in the scheduling engine guide.