Operations Management / Supply Chain Management

Module 06.02 Key Concepts: Capacity Planning and Management

Capacity must be considered at all levels of planning and execution for Operations Management.  This extends from high level Strategic Plan down through day-to-day operations.  At any level, if adequate resources are not available, the plan will fail.  Similarly if there are excess resources this usually means higher cost and lower productivity.  So, Capacity balancing is a key area of concerns for every function in Operations Management.

Capacity is defined in the APICS Dictionary as: “the capability of a system to perform its expected function or the capability of a worker, machine, work center, plant, or organization to produce output per time period. Capacity can be classified as budgeted, dedicated, demonstrated, productive, rated, safety, standard or theoretical.” APICS Dictionary 13th EditionThe text authors define capacity slightly different: “Capacity can be defined as the ability to hold, receive, store, or accommodate.  Strategic capacity planning is an approach for determining the overall capacity level of capital intensive resources, including facilities, equipment, and overall labor force size.”  As such, capacity impacts fixed costs and the ability to satisfy customer demand.Since capacity planning is relevant for all planning levels this encompasses three different planning views:

  • Long range: greater than one year
  • Intermediate range: Monthly or quarterly plans for the next six to 18 months
  • Short range: less than one month

Capacity is expressed in several different ways.

Design capacity is the maximum theoretical output of a system.  It is normally expressed as a rate.

Design Capacity is a product of the number of people and / or machines and the number of hours available.

Effective capacity is the capacity a firm expects to achieve given current operating constraints.  This is often lower than design capacity.

Two other factors are also relevant.

Utilization is the percent of design capacity actually achieved.

Utilization = Actual output/Design capacity

Efficiency is the percent of effective capacity actually achieved.

Efficiency = Actual output/Effective capacity

Capacity decisions impact all decisions of operations management as well as other functional areas of the organization.  Therefore, capacity decisions must be integrated into the organization’s mission and strategy.  Functional leaders must be able to: forecast demand accurately; match technology increments and sales volume; find the optimum operating size (volume); and build for change.  Each of these will be discussed below.  The big challenge is to manage capacity addition to provide the most effective economy of scale.
There are several options for managing imbalances between Demand and Capacity.
  • If Demand exceeds capacity
– Curtail demand by raising prices, scheduling longer lead time
– Long term solution is to increase capacity
  • If Capacity exceeds demand
– Stimulate market
– Product changes
  • If there is seasonal demands
– Produce products with complementary demand patterns
Additionally, the following operational approaches can be used to match capacity to demand:
  • Making staffing changes
  • Adjusting equipment: purchasing additional machinery; selling or leasing out existing equipment
  • Improving processes to increase throughput
  • Redesigning products to facilitate more throughput
  • Adding process flexibility to meet changing product preferences
  • Closing facilities

In Services, demand can be managed by appointments, reservations, and implementing the First Come – First Served Scheduling rule.  Capacity can be adjusted through addition of full time, temporary, or part-time staff.

 

Bottlenecks exist in all operations and represent a particular challenge.  You can see evidence of bottlenecks in all businesses and all aspects of daily life.  What happens when there is an accident on an Interstate Highway that takes one lane out of service?  Interestingly enough there is a backup of cars on the side of the highway directly impacted and usually a corresponding backup of cars on the other side of the highway as well due to people just looking at the accident.  What happens when you go to a restaurant during a busy time of day?  Yes, they give you a little vibrating-notification module and send you and your party to the bar or outside to wait.  What happens when you Customer Service and all the agents are busy?  Yes, you listen to music and commercials while you wait for someone to talk to you in person.  The same occurs in manufacturing work centers.  If there is a capacity over-load and work is still released, then backlog increases and queue time increases as well.  This extends lead-time to completion.  In fact, in batch operations and queue time is generally considered the longest component of total lead-time.

Each work area can have its own unique capacity. Capacity analysis determines the throughput capacity of workstations in a system.  As mentioned above, a bottleneck is a limiting factor or constraint a bottleneck has the lowest effective capacity in a system.

There are numerous publications, textbooks and formal programs / seminars that describe in depth the principles and steps used in managing bottlenecks and constraints in general.  These can be summarized in a very simple to say, but difficult to do, five step process.

The first step involves the identification of the bottleneck.  And, yes there are usually only one or two at most processes or work centers that represent a current bottleneck.  Bottlenecks are easily determined through the capacity planning processes we reviewed above.  They are also obvious from the backlog of work or high levels of work-in-process inventory in the work center / work area.  Walk around and see who has the most overloaded in-box.  Yes, you have located the bottleneck!

The second step requires assessment of the problem and finding creative short-term solution to either eliminate the bottleneck or at least improve its throughput.

The third step is to subordinate everything else to the bottleneck.. A bottleneck controls the entire throughput of an operation.  Work cannot proceed any faster through the whole operation than it can through the bottleneck itself.  All scheduling operations are done “forward” from the bottleneck and “backward” from supplying work centers.  This is a therefore a unique situation.  If a prior work center continues to produce at a higher level than the bottleneck can produce, work just gets “stacked up” and the bottleneck.   Bottleneck scheduling is sometimes referred to as drum-buffer-rope.  The bottleneck is the “drum” that paces all other throughput of linked work centers.  A “buffer” of work and / or materials is required at the bottleneck to insure that it never stops.  Buffers may also be required at final stock to ensure customer demand is met and at other forward work centers or raw material supply to assure that supplying work centers never miss deliveries to the bottleneck.  The “rope’ is the link from the bottleneck to the beginning of the process.  In this way the bottleneck controls the start of jobs.

The fourth step requires assessment of long-term needs, potential required investments in people, plant, equipment, materials, systems, etc.  It then requires analysis of the financial risks and rewards for the various options.  This is then followed by selection and implementation of the best option(s).

Managing bottlenecks is a never-ending process.  Once you eliminate one bottleneck, a new one will surface.  So, the fifth step is to begin all over again with the next “opportunity”.

Bottleneck Management includes many key assumptions / recommendations:

  • Release work orders to the system at the pace of set by the bottleneck
    • this is referred to a Drum, Buffer, Rope scheduling
  • Lost time at the bottleneck represents lost time for the whole system
  • Increasing the capacity of a non-bottleneck station is a mirage
  • Increasing the capacity of a bottleneck increases the capacity of the whole system

In order to be successful, Theory Of Complaints implementation requires a major shift in manufacturing mindset (paradigm shift).  The basic rule is that it is OK to not work (at a non-bottleneck) if there are no orders.  In addition, the Firm needs sound systems, education, top-management support, and a willingness to change.  If these are all present the long-term benefits will be realized.