I discovered a copy of Eli Goldratt's Essays on the Theory of Constraints on my bookshelf. I suspect it has been sitting there for five years. The Essays were originally published in the Journal on the Theory of Constraints (1987-90). While these may have been independent items when they were originally written, they mostly build upon one another - written sequentially for the journal, possibly.
I learned a few things in reading this book, and there were a number of familiar discussions of Theory of Constraints topics that helped me understand better. I'm not sure I'd suggest this book to someone new to TOC, as the style and writing aren't the most polished.
For people well-versed in TOC, Essays covers some familiar territory: Evaporating Clouds, the goal of a system, inherent simplicity, the role of science in explaining the world, the fundamental measures, the shift to a Throughput world... It also contains some entertaining true-to-life stories about companies struggling to improve their businesses and use some of the logic processes that are so important to TOC. In fact, the book might be entertaining for just the stories alone. I certainly want to know what happens to the fictional plant represented in the sequence of stories titled, "Looking Beyond the First Stage; Just in Time."
Below the fold, I go through more detail on each of the six chapters.
Chapter 1: Hierarchical Management - The Inherent Conflict
The inherent conflict in the hierarchical management organization is that "The total of local optima is not equal to the optimum of the total." The discussion had to do with the natural distortions that people make when evaluating information / passing commands. What causes this conflict? What is the solution? The direction of the solution has to do with the classic "evaporating cloud" technique that helps to highlight the assumptions and then break the dichotomy. In this case, there are some new rules that need to be put into place, as well as means to collect and distribute the right information. Explanations and discussions are built into the remainder of the book.
Chapter 2: Laying the Foundation; APOLOGIA
Here is the introduction of the Evaporating Cloud method for articulating conflicts (which Goldratt credits to a phrase from Richard Bach's Illusions). And it is used to describe the conflict from Chapter 1. The resolution Goldratt suggests is, "The objective of a local area is to contribute positively to the objective of the entire organization." This, in good logical style introduces several questions: what is the goal of the entire organization; how do we measure; how to contribute positively without massive amounts of information.
Going into a discussion of measures, the book takes a sidetrack into discussing cost accounting and the important ways in which cost is the wrong measure for judging both the local and global system. Of course, the discussion isn't a true sidetrack. Cost accounting is the way most businesses are still measured and make decisions today.
In the APOLOGIA section, Goldratt runs through an interesting discussion of the role of science in discovering classifications then correlations then cause-and-effect relationships. He suggests that the sciences all go through this cycle of growth with the "science" only happening when cause-and-effect relationships can be established. He then claims that management are only just getting into the cause-and-effect realm (with TOC being the cause-and-effect stage).
Chapter 3: The Fundamental Measurements
The fundamental measurements in TOC are Throughput, Operating Expense and Investment (Goldratt uses 'Inventory' here, which has been updated). The definitions are useful
- Throughput: The rate at which the system generates money through sales.
- Inventory: All the money the system invests in purchasing things the system intends to sell.
- Operating Expense: All the money the system spends in turning Inventory into Throughput.
This section also describes the way these measures can be used for other common calculations, such as net profit and return on investment. And the TOC measures of throughput-dollar-days (making sure you do the right thing) and inventory-dollar-days (making sure you don't do the wrong thing) are described.
Chapter 4: The Importance of a System's Constraints
Interestingly, this section doesn't really talk about constraints. It contains more discussion of the ills of cost accounting and cost allocation. Goldratt also acknowledges that just complaining about the idea of cost isn't helpful. Thus his proposal to use Throughput, Operating Expense, and Inventory as replacement measures. At some point Goldratt suggests (with hope) that the world will move away from the confusion of cost measures and into a world of Throughput Accounting. Sadly, that hasn't happened yet, 20 years later.
Chapter 5: How Complex Are Our Systems
Now Goldratt describes constraints: Market, Vendor, (internal) Resource, and Policy. The main focus is on how these constraints affect the system and how many constraints could really be limiting the system at a given time. The short answer is that while there may be many problems and issues in the system, there are only a very few constraints (usually one) that truly limit the ability of the system to grow throughput. This is the familiar-to-the-TOC community discussion of inherent simplicity.
Another way I've seen the inherent simplicity discussion is that a system is heavily interconnected set of processes and resources. But there is only place within the system, where you can make a change that affects the entire system. Still another way: there is only one "longest lever" to impact the system. If there are really multiple levers, then you either have multiple systems or you haven't found the core lever.
What about systems where the bottleneck appears to "float" from one resource to another? This is a sure sign of a policy constraint that is hiding an underlying constraint.
Chapter 6: The Paradigm Shift
What is the shared goal of many management concepts, such as TQM, Just-in-Time, Lean, Six Sigma, and Theory of Constraints? They all seek to address the need of companies to make more money now and in the future. It's just that they don't all use the same language, and the sound bite versions of their main focus misses completely that the larger goals are to ensure the organization is successful. Quality: ensure continued sales. Waste/Inventory: these programs are really about high service (improved sales). When reduced to their sound-bites, they tend to reinforce the traditional way of managing business, which is heavily focused on cost.
In Goldratt's view, the coordination amongst these approaches should be explained through T, I and OE, the key measures for Throughput Accounting. And the "paradigm shift" is to move Operating Expense from its position as king of the hill to its rightful place: third. Companies should focus on means to increase Throughput as their top priority. This leads to TOC's Five Focusing Steps, which take you through finding the constraint, exploiting the constraint, subordinating to the constraint, elevating the constraint, and watching for a new constraint. Investment is next (reduce it to a reasonable level), and Operating Expense is third. The other element of this paradigm shift, Subsystems within an organization are not independent of one another, so everyone must use the same metrics with the same focus on ensuring the system as a whole moves in the right direction.
Each chapter ends with a true story or a fictionalized version of a real-life situation. These were entertaining in themselves, particularly as they are written to show the logic that TOC applies to these situations. Given when this was written, all of the stories are manufacturing organizations. More recent TOC stories might include project management and the supply chain as well.
I was so interested in the story arc of the last three chapters that I was disappointed with the cliff-hanger at the end: just when they are getting ready to make a new leap in performance. What happened? How did it go? I will have to check with some friends to see if the rest of that story has been articulated anywhere.