Capacity buffers – a grossly underrated strategic concept

Available capacity is an issue for every organization or business. In order to deliver value any provider needs two different entities:  the appropriate capabilities to generate the value, and available capacity of each of those capabilities in order to react to the demand in an acceptable way.

Capabilities are easier to manage than capacity. The relatively difficult mission in managing capabilities is synchronizing effectively between the different capabilities. For instance, engineering capabilities are required to design a new product, and then production capabilities are required to produce.  The interface between the engineers and the production operators is far from being trivial.  These synchronization capabilities are part of the more generic requirements for management capabilities.  In order to generate throughput from new products additional capabilities are required, like purchasing, sales and finance and usually also IT.  All have to be managed to ensure the integration.

Capacity is the trickier part in the process of generating value. Answering the question “do we have enough capacity?” has to consider all the demand translated into load compared to limited available specific capacities.  Shortage of capacity, even of just one resource, is the ultimate cause for delays in delivery.  So, the practical focused question is:

Given the demand and the capacity limitations would our delivery performance be good enough?

A common complication of managing capacity is when on average there is ample capacity, but at specific times a bottleneck emerges causing very long delays in the delivery of value, probably far beyond the client tolerance time.

At such peak of load the threat of losing reputation leading to reduced future sales is very serious. Question is:

Is it possible to increase the available capacity at very short notice?

It is clear that if such short-term fixation of capacity is possible it is going to be relatively expensive. The point is that the damage of not paying that price could be much more expensive in the future.

Suppose the night manager of a hotel realizes that due to a mistake there is no availability of standard rooms for that night, but two clients with reserved rooms are expected to arrive from the airport.

This is a case where the clear damage of shortage of capacity of a specific resource is very high. It is pretty obvious that the hotel has to solve the problem – not the clients; otherwise the clients might sue the hotel causing more negative ramifications than just the compensation.

How can the hotel find more rooms?

If there are available upgraded rooms, like suites, then giving them to the clients is the first choice.  Note, these rooms are not part of the available capacity of standard rooms – yet, it is possible to use them for quickly increasing the capacity of standard rooms.  The next choice is to find equivalent rooms at another hotel, with similar characteristics.

Fluctuations in the market are the most vicious uncertainty that every organization has to be prepared for. Maintaining inventory is a common way to deal with this kind of uncertainty.  However, in most service environments, also in job-shops, holding inventory is not practical.  Inventory also protects only from peaks of demand for specific products, but only enough capacity protects from peaks of many SKUs at the same time.

How much available capacity of each type of resource (providing specific capability) to hold?

Maintaining very high level of internal capacity to cover all temporary peaks of load is a problematic business approach, creating considerably high OE and high pressure to generate very high T to cover for it. Yet, being able to respond to all demand is a necessary condition for maintaining business stability in the future.

The simple solution is distinguishing between two types of capacity: fully available, and capacity by request.  This leads me to define both:

Available Capacity: The periodical amount of capacity for specific capability that is regularly purchased by the organization and paid whether it is fully used or not.

For instance, an operator that is paid for 180 work hours every month. If that operator consistently earns 20 additional hours of overtime per month, then that amount of overtime should be included in the available capacity and this amount of overtime be included in the regular OE.

Capacity Buffer: Additional capacity that can be quickly purchased at relatively small amounts.

Capacity buffers are usually made of overtime, temporary workers, free-lancers, outsourcing and the use of resources that are primary used for different capabilities, like using the store manager as a cashier to reduce too long queues.

The capacity buffers protect the organization sales and reputation from temporary peaks.  The obvious cost of using the capacity buffers is the relatively high expense of using it.

Capacity buffers should be planned ahead of time and constantly managed! Looking for options at the last minute causes considerable damage in the long run.  Maintaining capacity buffers might cause additional costs.  For instance in order to add a special shift in the shop floor temp workers are needed.  It is unlikely to call temp workers that were not given work for several months.  Thus, to ensure having temp workers to call in a hurry they need to get minimum amount of work every month.

Thus, the global Strategy of the organization, looking to ensure the high reputation of the organization, should define the tactics of maintaining capacity buffers. The Strategy plan should detail the rules for maintaining the amount of available capacities and the rules for capacity buffers.

Even a capacity buffer might run out of capacity!

This is a considerable risk, because when the buffer is exhausted there is no buffer left and the damage is high as no one truly expects it to happen. It is necessary to monitor the capacity buffers using buffer management.

Implementing buffer management means the buffer is measured by the same measurement as regular capacity.  The most common unit of capacity is one-hour of a specific resource.  So, a buffer of operators in a manufacturing shop-floor could be 120 man-hours per week.  When additional 60 man-hours are used in one day – it means a penetration of 50% into the buffer.  Additional consumption of 30 man-hours means penetrating into the red-zone, which should generate a warning for the short-term, and feedback to the planning that the buffer might not be big enough.  A detailed periodical analysis of the performance of the capacity buffers has to be in place, both for better understanding of the flexibility level the organization has to maintain to satisfy the market requirement, and to keep reasonable control on the operating expenses.


Published by

Eli Schragenheim

My love for challenges makes my life interesting. I'm concerned when I see organizations ignore uncertainty and I cannot understand people blindly following their leader.

5 thoughts on “Capacity buffers – a grossly underrated strategic concept”

  1. Eli, nice article.
    Probably capacity buffers can be planned on every aspect of the organization. For example: Procurement developing providers

    How much available capacity of each type of resource (providing specific capability) to hold?


  2. Finance, having cash available from banks, Support areas having third parties ready for managing load, Maintenance with a list of experts ready to come, and so on.
    From time to time some areas of the organization experiment relatively high load and capacity buffers play a fundamental role for ensuring that the pace of the system remains stable.

    Eli, you wrote: How much available capacity of each type of resource (providing specific capability) to hold? I think that the answer to that deserves a third article!


  3. Also, in operations there must be a relationship between the strategic concept of capacity buffer and the concept of Protective Capacity.
    When a non constraint suffers for a full consume in the protective capacity, then is clear that the activation of the capacity buffer is mandatory. If the resource provides material to a CCR then the CCR will show less load in its bank of work, in the other way the resource provides material to finish goods, then the wip increased and consequiently lead time increases. So, the impact of the capability of the organization to quickly activate the capacity buffers can be exposed through the view of the planned load, isn’t it?


  4. Gustavo thanks a lot. I’ll write a third article on making the decisions between maintaining enough available capacity, having partial available capacity and then enough capacity buffer and fully rely on capacity per request.

    In my first post on the topic: Protective Capacity and Capacity Buffer I have explained that the capacity buffer serves as protective capacity. It could be that all the protective capacity is part of the capacity buffer, or there is some protective capacity from the available capacity and some additional protective capacity in the shape of capacity buffers.

    You are also right that the basic definition of the Planned-Load do not consider the capacity buffer, because any use of the capacity buffer causes delta-OE. However, when a decision is made to use some additional capacity, I’d like to update the planned-load to reflect the full impact of the additional capacity.

    Liked by 1 person

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s