Written by Dr. Alan Barnard and Eli Schragenheim
We both encountered the name of Prof. Herbert Simon, long before we met Dr. Eli Goldratt. Prof. Herbert Simon (1916-2001), a recipient of the Nobel Prize for Economics in 1978, was an American political scientist, economist, sociologist, psychologist, and computer scientist. Prof. Simon was among the founding fathers of several of today’s most important scientific domains, including artificial intelligence, information processing, decision-making, problem-solving, organization theory, complex systems, and computer simulation.
He coined the terms bounded rationality and satisficing.
Bounded rationality is the idea that, when we make decisions, our rationality is limited, not only by the inadequate information we have available and/or inadequate knowledge to predict the outcomes of our decisions, but also the cognitive limitations of our minds, and the limited time available to make these decisions.
Simon coined the term satisficing (a combination of satisfy and suffice) to describe the heuristic we likely use when having to quickly make difficult decisions with inadequate information and/or knowledge.
Simon often said: “people are not optimizers they are satisficers” – we seek a satisfactory solution rather than an optimal one. When faced with a challenging problem or decision, we search for a solution that satisfies our pre-defined criteria to a sufficient level. When such a solution is reached there is no need to continue searching – we have found a solution that is good enough!
We both rediscovered Simon’s incredible insights, when we recently started doing research on the limitations managers are confronted with, which Theory of Constraints and its applications can help diminish or even eliminate. These are the limitations imposed by complex and uncertain situations as well as by conflicting objectives in solving problems and making decisions.
Below are three of the highlights we found during this “rediscovery”.
In 1971, the world was just at the beginning of the huge advancements in information technology and the exponential growth in the access to information. Yet, Prof. Simon already had the foresight to warn us about one of the major negatives of the increased access to more and more information.
In a public speech he gave in 1971 he warned:
“The wealth of information means a dearth of something else: a scarcity of whatever it is that information consumes. What information consumes is rather obvious: it consumes the attention of its recipients. Hence a wealth of information creates a poverty of attention and a need to allocate that attention efficiently among the overabundance of information sources that might consume it.”
And he went further. In his 1973 paper titled, “Applying Information Technology to Organizational Design”, Prof. Simon wrote:
“The information-processing systems of our contemporary world swim in an exceedingly rich soup of information. In a world of this kind, the scarce resource is not information; it is processing capacity to attend to information. Attention is the chief bottleneck in organizational activity, and the bottleneck becomes narrower and narrower as we move to the top of organizations.”
Dr. Goldratt gave similar warnings. Firstly, in the Haystack Syndrome, he warned about the importance to differentiate between data and information (the answer to the question) and the need to build true information systems that deliver only the information relevant to managers for making important decisions. Later, he also shared his insight that the ultimate constraint in any organization was Management, especially Top Management’s limited Attention.
Goldratt explained in “Standing on the Shoulders of Giants” that he simply advanced the work started by Henry Ford and Taiichi Ohno, realizing that to improve flow you need a practical mechanism to prevent overproduction – producing things that are not needed, or at least not needed now.
To our mind, Dr. Goldratt also advanced the work started by Prof. Simon, by outlining practical mechanisms for helping managers decide what to focus on (and as importantly, what not), to better exploit and not waste the scarcest resource in any organization – management’s limited attention.
Considering the growth in Big Data, what are the implications of these warnings on managers’ problem solving and decision making today?
Will it really help to provide the step-change in managers and domain specialist’s ability to improve the quality and speed of their decisions?
And, even if it did, will it be sufficient?
And the third insight: A citation from Prof. Simon article entitled “Making Management Decisions the role of intuition”:
“What all of these decision-making situations have in common is stress, a powerful force that divert behavior from the urgings of reason. They are examples of a much broader class of situations in which managers frequently behave in clearly nonproductive ways.”
We, Alan and Eli, are deeply interested in the impact of fear on the way managers make decisions and manage organizations. We all fear being blamed for our decisions and actions. So it is safer not to act. But we also fear missing something (in the ocean of data) and being blamed for not acting. Whether we do (act) or don’t, we are damned.
Fear results in stress. And we know that when people are under stress, they often freeze-up, not doing something they should, or they over-react, doing something they should not. When under fear-induced stress, we often act in irrational ways.
Prof. Simon also frequently warned against excessive fear of unforeseen consequences. He advised that the best way to overcome such fears was to experiment and to see what happens.
We have four questions regarding Prof Simon’s concepts and its implications on managers today.
- Are decisions made within the organization also aimed at satisficing rather than at optimizing? Our reason for asking that question is our observation that organizations seem to impose the value of optimization on decisions, and by that, almost force managers to look beyond satisficing, leading to be considerably less focused, and resulting in decision delays and/or errors.
- To what extent today, do the fear of being blamed for taking the wrong decisions/wrong actions, still cause many of the avoidable decisions errors and delays?
- To what extent today, do the fear of missing something important in the data, still cause managers to look at too much data / too many measurements, causing distractions which in turn, waste management attention and also result in decision errors and delays?
- Why, assuming there is consensus on the timeless importance of the above insights from Prof. Simon, a Nobel Prize recipient, have many others not followed and continue this important work?
The answers to these questions could hint on why the awareness and adoption of TOC is still much lower than what we have expected…
15 thoughts on “From a TOC perspective: Paying tribute to a Great Pragmatic Thinker”
Thank you for this excellent article to both authors. Well done! Due to my current studies in the field of data science, I want to reflect on two of your questions:
“Considering the growth in Big Data, what are the implications of these warnings on managers’ problem solving and decision making today?
Will it really help to provide the step-change in managers and domain specialist’s ability to improve the quality and speed of their decisions?”
There are powerful tools in the area of Big Data, which are capable to generate a huge amount of derived information. But from a management decision point of view they can not help as long as one basic question isn’t answered by management:
“What is the question you are trying to answer by big data analysis?”
As long as a manager isn’t able to give an answer, Big Data Analysis can only deliver useless information. There is no difference compared to small data sets.
But who knows? Perhabs one day we can help management with artificial intellligence (AI) based on machine learning algorithms to make better decisions — the TOC way — .
Relevant information is defined by the decision at hand. this means any questions for Big Data, of any other data source is determined by the decision or for validating or invalidating a hypothesis, which would lead to many decisions.
I have some tiny problems to understand your remark regarding the point that Big Data will lead to many decisions. Our production and or supply chain systems are already producing data as a side product. What is wrong to combine data from several sources to gain deeper insights about the entire system as relevant information for decisions at hand? What is wrong to validate or invalidate a hypothesis? All this information is useful as long as it delivers answers to the right question.
Juergen, sorry for the misunderstanding. What I meant was the positive potential of Big Data. When Big Data is used for validating or invalidating a hypothesis the revealed paradigm would lead to better understanding of the cause-and-effect and have very positive impact.
My concern is that the analysis has to be triggered by the decision or by a hypothesis, which means every analysis is focused on a specific need and not describing huge amount of trends, which consume management attention, but do not improve any specific decision. It is so easy to lose focus.
Excellent article!!! We, TOC practitioners must be careful about to over-sophisticated solution apparently developed from the Theory Of Constraints but contradicting the essence of good enough!
Powerful and insightful message. The wise words of Prof Simon that attention is the chief bottleneck of management, is so apt in a world that demands selflessness and servitude. Those words will resonate with me profoundly. Alan, I have always held you in high regard and would like to engage you on the subject.
Amazing article! So many times I have seen technology is assessed without the context of business. Keep up the great work Eli!
Great article Eli & Alan. Thanks.
Let me start with an answer to question #4:
Because there is so much surplus data (not useful to managing the overall flow but great at making local areas potentially more efficient) that the fear of missing something keeps manager looking at more of the surplus type instead of acting (or not) on the key information.
Back to #3.
I think it is the biggest problem facing management. Everywhere I go, management attention is exhausted. Most people feel like they are goofing off, if they don’t stay active and display their activity, especially managers. The exception is when showing off that they don’t have to work, by vacationing (I am writing this while on vacation – ironic) and taking long lunch breaks or coming to work after everyone else. Yet, the showing off exception makes them all the more stressed out when they are working, because they feel behind.
Back to #2.
The search for the perfect decision – one for which one cannot be blamed is rampant, especially in companies big enough or old enough to have layers of politics which support the status quo, usually from past glory days. Young and small companies tend not to have this characteristic. People want to take a chance on success, only limited by their own personalities as to when they pull the trigger.
I don’t know about #1. I have definitely seen satisficing decisions. There is lip service, sometimes expensive lip service, towards optimizing. This is the hope for super expensive ERP, MRP and other computing systems. Even if the computing tool is only expensive with respect to the time and though put into its programming or loading it with appropriate data, the hope is for an optimal answer. But, I have never seen such output, even when I wrote the ERP system myself!
I see managers constantly compromising on their answers for some excuse.
I am not writing about a computer system on a certain price level. My point are the data coming from ERP, MRP, Web, sources along the supply chain, etc. The combination of these data followed by a powerful analysis will hopefully lead to deeper insights of my system(s). I am aware about lip service. Again my point are the opportunities we will get in future to uncover undesired effects like lip services with (big) data analysis.
More on #3: At the same time, as people feel the need to know everything, there are things they purposely but surreptitiously ignore. For examples, I see people pretend to ignore critical branches of thought, especially if they can realistically claim that other people ignore them too. Likewise, when reading something they think is really important, I seldom see people searching out sources footnoted in this self-assessed important reading.