| 01 March 2010
If we look at the number and life span of “new” approaches to operations management (and how managers choose among them) over the last few decades, we might conclude that it is more like following the fashion trends than a scientific methodology.
The authors of this extremely valuable book, Factory Physics ® (Hopp & Spearman. 2008. New York: McGraw-Hill/Irwin) present a comprehensive, yet accessible, text on operations management. They show that, despite the weaknesses of any one method, the greatest value-add lies in knowing which methods to use in your business. In short, they lay out a structured and quantitative approach to operations management. To illustrate the value of this approach, ask yourself if you see your company in any of these scenarios:
- Your production planners spend their days firefighting, despite having expensive software that is supposed to help them.
- You would like to put your Work-In-Process (WIP) and Cycle Time (CT) metrics in context, but question the value of comparing (“benchmarking”) your company's performance to others.
- You suspect your WIP is too high, but you are concerned that a Kanban system in your environment would be too complex.
I will describe how each of these dilemmas can be addressed.
No More “Flavor of the Month”
As often happens, individuals with insight work to determine what approaches (perhaps new approaches) will make their company successful, given their objectives, products, processes, and competition. They come up with something that works, and their success stories are publicized. Others try to apply the methods, often to situations for which they are ill-suited, resulting in failure. Then the failures are publicized, and the methods become less popular and people wait for the new “Flavor of the Month”, normally brought in by some consultants.
The authors of Factory Physics ® help practitioners avoid all this by developing a “science of manufacturing”. Their primary goal is “to provide the reader with an organized framework from which to evaluate management practices and develop useful intuition about (their) manufacturing systems.” Beginning with a historical overview of operations management, they highlight the contributions, as well as the flaws, arising from each major development in the field. Armed with an improved understanding of the traditional operations management tools, as well as some new concepts, the rest of the book describes a framework for using these approaches to manage a manufacturing operation.
To demonstrate the value of this approach, I will address briefly what the book has to say about the three scenarios above. Each represents an insight that I believe is significant, but is not well known in the operations management community.
Turning Firefighters into Planners: The Central Flaw of Material Requirements Planning (MRP)
MRP is invaluable for coordinating large numbers of materials purchases to a Master Schedule. Its expanded version (Enterprise Requirements Planning or ERP) seems complex, but the central calculations are easy to understand. Nonetheless, many companies struggle with their MRP systems, constantly revising purchase orders and production schedules.
The problem lies in a fundamental assumption made by all MRP systems, which is all production lead times are fixed. In reality, they are highly variable. High capacity utilization will increase lead times, and low capacity utilization will decrease them – which apply to the lead times from your vendors and for your internal production lead times. The resulting errors lead to constant rescheduling, but there is hope. Factory Physics ® describes how to minimize this while preserving the benefits of your MRP system, ultimately reducing the occurrence of firefighting, improving the reliability of order completion dates and, in turn, on-time deliveries.
Realistic Goals: Being Lean Without Starving
Some Lean authors imply that having no inventories ("Zero Inventories") should be the goal. However, in reality, with no inventories of WIP there will be no production. So how “lean” should an operation be?
Defining “performance” as the ability to produce more output with a given level of WIP, a method of "internal benchmarking" is described, where a company can determine the "Best Case", "Worst Case", and “Practical Worst Case” performance that can be expected from their processes. The actual processing times required for the process steps are kept constant across all three cases. However, each case assumes a particular approach to manage the process, and this leads to extremely different (theoretical) performance. A manager can then compare the actual process performance to the three calculated cases.
For each calculated case, two graphs are developed. The first shows throughput as a function of WIP, and the second shows CT as a function of WIP. Once these curves are created, a manager can compare the actual performance of an operation to the curves and prioritize efforts towards those that show the most opportunity for improvement. These curves also highlight the level of WIP below which throughput will be severely reduced. Since this is a custom calculation for each individual operation, it avoids the mistake of comparing pears to apples.
Pulling Without Kanban: The CONWIP Approach
Most books and articles on Pull systems focus on describing the Kanban mechanism, rather than the root cause of why Pull systems are helpful. Factory Physics shows that the main contribution of Kanban systems is that they limit the WIP inventory levels and, in turn, cycle times. This WIP reduction is a direct result of all Pull systems, not just Kanban.
Contrary to some authors, setting up and maintaining a Kanban system with the correct number of cards at each step of the operation is not trivial. A simpler Pull system is described, called CONWIP (for "Constant WIP"), which achieves most of the benefits of a Kanban system but with much less effort. This leaves operators and planners free to focus on their work, rather than struggling to maintain the system.
Putting It To Work
As an experienced Industrial Engineer, I have seen first-hand much of what is discussed in the text, both the good and the bad. The three concepts that I have highlighted are just the tip of the iceberg; every section has similar valuable insights. It is true that some sections utilize significant mathematical and statistical calculation. However, the practitioner is not left behind: the proofs are very clear. And if need be, they can be skipped and the concepts will still be understood. In fact, many useful “intuition-building exercises” are included to help the reader solidify his understanding.
So, whether implementing Lean methods, searching for low cost solutions to increase capacity, or developing an overall operations strategy, I can bring forth the correct methodology to support every unique situation. This approach is especially helpful in defining what “improvement” means, and in setting expectations and priorities. I would be glad to apply my knowledge to your operation and help you get started.
FACTORY PHYSICS is a registered trademark of Factor Physics Inc. All rights reserved.