Production capacity vs. demand and utilization rate

In one line

Manufacturing capacity is what the plant can produce in a period; demand (what is sold or promised to be sold) is what has to be produced; and the utilization rate tells you what part of the capacity is being used.

What it is

The UNGS book distinguishes two capacities. Theoretical capacity is the company's potential ability, the one that comes from the installed technology level. Real capacity adds the dynamic ability: the capacity of the company's social system to adapt to that technology and to manage itself. In general the real capacity is lower than the theoretical one.

In a flow production line, the capacity is set by the slowest workstation. The book works this out with an exercise about a croissant factory: the cycle time of the system is the same as that of the workstation that takes the longest, and the capacity is the theoretical number of units that can be made in the available time:

Capacity = available time ÷ bottleneck cycle time

Required utilization = demand ÷ capacity

Manufacturing vs. selling

If what is sold (or promised to be sold) exceeds what can be manufactured, the result is failure to deliver; if what can be manufactured exceeds what is sold, the plant has idle capital. Both situations are costly, which is why it pays to look at them together.

The snack factory in Quito from the Universidad Tecnológica Equinoccial article is an example of the first: it had a customer service rate of 89.58% and failed to sell 1,340 bags because of process constraints. Its fryer had a capacity of 1.20 kg per minute, about 576 kg in an 8-hour shift, but it processed 363.3 kg per day: a fryer capacity utilization of around 63%, with the constraint working about 5 of the 8 hours. The article's proposal (better scheduling, smaller batches and shorter setup time) projects a production about 21% higher and 100% fulfillment of orders. And the article anticipates the second: once the manufacturing constraint is resolved, the constraint moves to the market, and the finished-product inventory has to go on sale. It is the logic of the theory of constraints.

Calculate your capacity

Real example: croissants

The UNGS book's times per dozen are 3 minutes of mixing, 2 of shaping, 6 of baking, 3 of cooling and 1 for changing trays and service. The bottleneck is baking: 6 minutes per dozen, or 10 dozen per hour, which over 8 hours gives 80 dozen per day. If demand is 60 dozen per day, the required utilization is 75% and the plant has enough capacity. With 100 dozen per day it is not enough, even though the other workstations have capacity to spare.

The improvement the book proposes is to add another oven: baking capacity goes up to 20 dozen per hour, but mixing and cooling also yield 20, so the constraint ends up shared among three stages. One clarification: the book's printed answer swaps the baking values (it says 10 minutes per dozen and 6 dozen per hour); here we use those from the problem statement, which are consistent with its own solution.

Utilization of machines and people

Another way of talking about utilization is the fraction of available time that a piece of equipment or a person is actually producing. Niebel and Freivalds measure it with work sampling, and give a case: in a heavy machinery shop, management estimated that the real cutting time of the 14 machines should be around 60% of the workday to meet the budgets. After about 3,000 random observations (3,024), the real cutting time turned out to be 50.7%; the remaining time included 9.6% for setup and 10.8% for tool handling, which pointed to where to improve methods in order to increase cutting time.

Template to use

Workstation	Cycle time (min/unit)	Capacity per shift	Demand per shift	Required utilization
(workstation 1)		available time ÷ cycle time		demand ÷ capacity
(workstation 2)
(workstation 3)

The workstation with the lowest capacity is the bottleneck. Next to the table, it is worth noting the real available time (subtracting scheduled stoppages) and, if possible, the OEE of that equipment.

So you do not have to build it by hand, there is a takt time and capacity per workstation spreadsheet in Excel with the formulas already in place: it calculates the capacity and utilization of each workstation, the bottleneck and whether the line can meet the demand. The values it comes with are examples and need to be deleted.

Benefits

• It lets you say "yes, we can make it" or "no, we are short by this many units" with numbers, before committing to an order.
• It points to where to invest (or where to improve) to increase the capacity of the whole line.
• It separates the capacity problem from the utilization problem: they have different solutions.

Limitations to keep in mind

• It uses measured cycle times: if they are wrong, the whole calculation is wrong (see time study).
• A utilization close to 100% is not a goal: it leaves no margin for stoppages, product mix changes or demand peaks.
• It does not consider quality or stoppages: OEE is used for that.

In summary

Capacity is what the bottleneck allows; demand is what has to be produced; utilization is the relationship between the two. Looking at both capacities, the one to manufacture and the one to sell, avoids the two extremes: promising what cannot be delivered and keeping an oversized plant. Related: takt time (the pace that demand requires) and line balancing (spreading the work to even out the workstations).