Manufacturing Breakthrough Blog

Local Optimum Versus Global Optimum Part 1

Friday December 2, 2016

In my last several posts we spent a lot of time describing the standard cost system, the fundamental decision-making technique used by many manufacturing companies.  Essentially, we uncovered the limitations of this system when looking at purchasing decisions.  In the next several posts we will dive into the conflict that exists between “local” and “global” performance measures, especially when operational decisions must be made in real time.  One of the most important functions of performance measures is to motivate the right behaviors, but the standard cost system encourages dysfunctional behaviors.  I will also recommend what I believe is a superior system to help you make better decisions, effectively evaluate and enhance organizational performance and encourage people to take appropriate actions.  The bottom line is that managers need a set of measures to help them make decisions that enhance the performance of the “global system” rather than local improvements to isolated parts of the system.

If you ask many managers what the goal of their organization is, it’s not uncommon to hear things like “to provide high-quality products at competitive prices; to provide better customer service; to capture market share; to reduce costs, etc.”  In reality, the goal of all for-profit companies is to “make money, both now and in the future.”  With this goal in mind, every management decision, from all parts of the organization, must be directed toward helping the organization reach this goal.  The problem that I have seen in many companies is that most companies lack the appropriate guidelines to help managers achieve this goal.  So let’s focus our attention to the real role of performance measures as they apply to making decisions that enhance the performance of the organization as a whole and not just parts of it.

In a typical company, during the first part of the month, the actions of the entire work force are heavily influenced by standard cost accounting performance measures which primarily have a “local” focus.  Measures like manpower efficiency and/or equipment utilization, as they apply to specific machines, the workers themselves, work centers and departments are quite common.  These measurements stress the standard time to process a part as well as the cost to produce one unit at each operation.  As Srikanth and Umble [1] explain, the supervisors act as if they are wearing blinders as they strive to meet the established time and cost standards, without regard to how their actions affect the rest of the plant.  Individual departments are working in virtual isolation, each striving to meet their own efficiency standards, can have a devastating effect on downstream operations.  Perhaps the most glowing example of this is a supervisor running an excessively large batch of a particular part in order to economize on setup time and drive efficiencies higher.  Of course, what happens is that we most likely find excess work-in-process (WIP) of some parts and shortages of others.

The excess WIP occurs because of the large batches result in more units of the part being produced than are required in the short term to satisfy an order.  Shortages occur because a lengthy production run of any one item causes a delay in the timely production of other items that are required at downstream operations.  The performance metric looks good, but some operations may develop huge backlogs of work while others may have to shut down while they wait for parts to arrive.  In addition, some supervisors may also create bottlenecks by utilizing their most efficient equipment so as to keep their efficiencies high.

As the end of the reporting period approaches, with shipments now behind schedule, general financial measures (e.g. monthly shipping goals) for the plant as a whole take center stage and the plant manager intervenes.  The workers are now told to do whatever it takes to meet the shipping targets for the reporting period.   Because the shipping targets are in jeopardy, metrics like efficiency and utilization are temporarily abandoned in favor of the “global measures.”  As a result, production workers are forced to work overtime, expediters take actions to split/overlap batches, the schedule changes and a host of other panic-driven actions are taken so that shipping targets (a global measure) can be met.  The problem is, when the new reporting period starts, this vicious cycle repeats itself, month after month and year after year.

This example demonstrates just how inefficient and nonsensical this mode of operation really is.  This constant shifting back and forth between working on achieving local optimums until the end of the reporting period enters the picture and then shifting toward meeting global optimums is ridiculous. The problem is that for most of the month individual actions are taken to satisfy local performance measures when the focus should be on satisfying the company’s goal of making money.  Clearly, the reason these traditional cost-based procedures are so misguided, and cause dysfunctional behaviors, is because they reinforce the idea of maximizing the performance of individual subsystems rather than optimizing the performance of the total system.  It should be obvious that in order to improve the overall system, you must make sure that your performance measurement system supports the overall goal of the organization and not subsections of it.

Next Time

In my next post, we will continue our discussion on local versus global optimums.  As always, if you have any questions or comments about any of my posts, leave me a message and I will respond.

 

Until next time.

 

Bob Sproull

 

References:

[1] L. Srikanth and Michael Umble,Synchronous Management – Profit-Based Manufacturing for the 21st Century, Volume One – 1997, The Spectrum Publishing Company, Wallingford, CT

Categories :

Leave a Comment (Please Log In First)