An advantage does not "cancel-out" dis-advantage, the disadvantage remain until 'corrective action' (gets rid of the known cause of a problem) are taken.
approach
p.202, p.203
Note that in choosing house A the manager knew why and on what grounds he was doing so. He did not take the "cancel-out" approach used by some managers in decision making. In this approach, ([in this approach]) the assumption is that an advantage cancels out a disadvantage so that things even up. This is not so.
([pause])
If there is a disadvantage attached to an alternative, finding an advantage does not get rid of it. Once the decision is made, the disadvantage will have to be lived with until it is removed by corrective action of some sort.
([pause])
The only safe way to deal with disadvantages in decision making is to recognize them and to keep them visible before one throughout the process. A final decision or course of action can then be made in full knowledge of the disadvantage rather than by glossing over defects and hiding them.
([pause])
Having all the assessments that enter into a decision visibly set forth is a major advantage in itself. For one may readily go back to reexamine the judgements made and consider corrective actions that can be taken to improve an already good alternative.
1965 by Charles H. KEPNER and Benjamin B. TREGOE
(The Rational Manager : A Systematic Approach to Problem Solving and Decision Making, Charles H. KEPNER, Benjamin B. TREGOE, 1965, p.202, p.203)
____________________________________
The Rational Manager
A Systematic Approach to Problem Solving and Decision Making
Charles H. KEPNER
Benjamin B. TREGOE
(Kepner-Tregoe Analytical Methods, root cause analysis, statistical process control (SPC), total quality management(TQM), six sigma)
Introduction
By Perrin Stryker
This book needs no introduction to some 15,000 experienced managers who by 1965 had gone through the training programs developed by Drs. Kepner and Tregoe. It is safe to say that every one of these managers made the same startling discovery that his own private system for handling problems and decisions simply did not work very well and often did not work at all. This is, unfortunately, a discovery that few managers seem to make in the course of their own careers, and they fail to make it largely because the re-education and improvement of their reasoning habits have not been considered
necessary.
Yet the cost unsystematic and irrational thinking by managers is undeniably enormous. If he wants to, any good manager can easily recall from his experience a wide assortment of bungled problems and erroneous decisions. As an executive of a large corporation long honored for its good management once said to me, "The number of undisclosed $10,000 mistakes made in this company every day makes me shudder." However, like others in management, this executive did not think his subordinates could be trained to think more clearly about problems and decisions; he remained inarticulate about his own thought processes and did not seriously question his habits and methods in handling problems and making decisions. This is understandable. It is much easier for a manager to study things like finances, materials, and markets than it is for him to turn his mind upon his own reasoning processes.
Moreover, clear concepts about the reasoning processes used by managers are still scarce. In recent management literature a lot of attention has been paid to the difficulties involved in handling problems and making decisions. (Brief descriptions of some of these studies will be found in bibliography.) But nearly all the theories and systems proposed seem to be either obscurely complex or obscurely thin and superficial, and they commonly confuse the processes of problem analysis with those of decision making. The arrival of computers and data processing appears to have compounded the confusion. On the one hand, managers are urged to apply computers to their business problems, while, on the other hand, they are frequently reminded that such mechanisms and procedures cannot make their decisions for them. The reminder is essential, for, no matter how many computers he uses, the manager himself still has to know how to reason clearly about problems and their possible solutions.
There is, of course, more involved in being rational manager than the ability to think through problems and decisions logically and systematically. For one thing, a manager needs good judgement to make good decisions, and this capacity is itself a compound to experience, values, and innate abilities which may dictate courses of action that are not neccessarily the products of strictly logical reasoning. Setting objectives and policies similarly involves considerations that may not be reasonable from one or another viewpoint. But the capacity to reason systematically is unquestionably a basic neccessity for any manager who hopes to manage well. Successful managers have developed this capacity through experience. But the first chapter of this book shows that even experienced managers are
surprisingly inefficient in the ways they go about handling problems and decisions.
It was this inefficiency that attracted the attention of the authors during the 1950s while they were working for the RAND Corporation in California. For several years up to 1958 Kepner, a social psychologist, and Tregoe, a sociologist, had spent most of their time working with advanced systems of defense for the Air Force, and this "ivory tower" occupation eventually got them thinking about the ways people use information in a highly automatic data-processing system. On their own time they studied the effects of automation in industry, and from there it was only a step to the processes of decision making. They soon saw that a great many business decisions were often wretchedly made and extremely costly. They were convinced that there must be some principles and basic techniques that would improve this managerial performance. From their work at RAND they knew that new directions would have to be followed if they were going to develop the underlying concepts of decision making. So they decided to leave their jobs and undertake this research entirely on their own.
The progress of their research is instructive. First, they reviewed the literature on problem solving and decision making, looking for techniques and concepts or principles that might explain the difference between good and poor decision making. They found bits and pieces, but precious little that they considered useful. Then they examined the internal workings of an organization from policy level to accounting procedures, looking at its complete operation. But these business details did not help in finding concepts that could be used in solving problems. Sitting in Tregoe's garage they spent hour after hour trying first one idea or techniques and then another, but nothing worked out to their satisfaction. Then one day they completely reversed their attack and decided to start with a problem of a company and work backward through the process of solving it, dissecting the thought process involved at each step.
This engineering approach produced entirely different results. They developed a set of ideas based upon what a manager has to do to solve a problem in real life; then, to help make these ideas visible, they developed a simulation of an imaginary company which they call "APEX." The first problem they posed for research was a series of customer complaints about one style of APEX's screen doors. They chose this problem because Tregoe happened to be looking at his own screen door as they sat in his garage, and they figured that some trouble with a door would be an easy thing to visualize.
As they began to think through the detail of this screen door problem, they began to expose some of the common confusions of problem solving; they saw, for instance, that it was no use to advise a manager to "define the problem" before he had even identified which was the most important or urgent problem he had to deal with. Nor was it fruitful for a manager to ask why a problem had occurred before he knew exactly WHAT it was he had to explain. By such pragmatic thinking they followed through the processes that would be involved in reaching a correct solution, and they backed up their findings with more field research. They spent six months asking managers in a variety of businesses about the steps these managers actually took in solving typical problems and making decisions about them.
This research produced many fresh concepts. While the concepts in existing textbooks were based on one discipline
or another, those of Kepner and Tregoe were eclectic and did not fit with even bits and pieces of the established works on
problem solving and decision making. For example, they developed the concept of a problem as a deviation from a standard, and the concept of cause as unplanned, unexpected change, and only later learned that Professor Herbert A. Simon was at about the same time independently arriving at similar concepts in his computer research on problem solving theory. For their analysis of problem solving and decision making, Kepner and Tregoe eventually developed fourteen concepts (summarized in Chapter 3) and then moved on to the process of potential problem analysis which is dealt with in the last chapter of this text. The whole progression took nearly five years to develop to its present stage, and further developments and refinements will undoubtedly follow.
The validity of these concepts and procedures has already been established in the most convincing way, by practical applications on the job. Some of these applications, described in Chapter 9, demonstrated the flexibility that managers have used in adapting the ideas and techniques to special situations. It is also notable that training in the use of the concepts and procedures has become a regular part of management development in some of the leading United States corporations, including General Motors, Ford, Du Pont, General Electric, Honeywell, and IBM.
Managers themselves will probably find the training methods devised by Kepner and Tregoe and their associates to be as impelling as the concepts. The authors present a full explanation of their teaching methods in the appendix, but I would like to emphasize here that this training provides a kind of experience that managers rarely, if ever, get. In this three part training method, the first part--study of the concepts--is familiar enough and parallels the kind of vicarious management experience gotten through lectures, discussions, audio-visual and case study methods. The second part--intensive practice of the concepts and procedures in a simulated business situation--is also not entirely new, though in its sophistication and intensity it differs from in basket techniques, workshops, and role-playing sessions. But the third part of this training--the feedback sessions on actual performance--is not duplicated anywhere else to my knowledge. For these feedback sessions provide the managers with an immediate and detailed critique of their actual performance and show them HOW they went about solving problems and making decisions, WHAT was wrong or might be improved, and HOW they can revise their performance. By way of contrast, managers playing business games with computers are in effect simply guessing the computer's philosophy of business and management; they do not learn how they went wrong and specifically how they can improve their decisions--and, of course, such games teach nothing about the analysis of problems.
The research and training methods developed by Kepner and Tregoe carry certain inevitable and significant implications for anyone interested in the processes of management. Their work clearly shows that problem analysis and decision making are management acts that should be consciously and systematically performed, and if necessary, recorded. The idea that a manager should be conscious of exactly what he is doing while he is managing may not sound revolutionary, but the fact is that such management is seldom found in industry today. The absence of conscious, systematic problem analysis and decision making is not only responsible for inefficiency and waste; it is also responsible, in large part, for the general neglect of two of the most important management functions: the setting of clear objectives, and the setting of clear performance standards for personnel.
Perhaps the most significant implication of the concepts and procedures described in this book is that they anticipate the kind of manager that will be needed in the future. As the authors have pointed out, the continuing increases in technology inevitably mean that managers will know less and less about the skills and knowledge of those they are managing, and will have to depend more and more on their ability to manage the operating techniques of those reporting to them. And in order to manage the way a subordinate handles problems and decisions, a manager has to know how to ask just the right questions. It is precisely with this critical management skill that this book is ultimately concerned. Such skill cannot, of course, be acquired overnight, and one reading of this book will certainly not develop it, but in my judgement there is no better place to start.
CHARLES H. KEPNER, a graduate of Fresno State College, holds an M.S. degree in psychology from University of Oregon and a Ph.D. in social psychology from University of Michigan. He spent several years in contract research in Canada in the fields of mental health, conservation, and medicine,
and was research director for the Royal Commission on Agriculture and Rural Life. Following two years in military training research at the RAND Corporation, he left to join Tregoe and set up their organization to study problem
solving and decision making.
BENJAMIN B. TREGOE, a graduate of Whittier College, holds a Ph.D. degree in sociology from Harvard, and has lectured at University of Southern California and at University of California at Los Angeles. For three years he was field training consultant to several Air Force installations, and was doing systems research at the RAND Corporation before joining Kepner to establish KEPNER-TREGOE AND ASSOCIATES, INC.
1965 by Charles H. KEPNER and Benjamin B. TREGOE
(The Rational Manager : A Systematic Approach to Problem Solving and Decision Making, Charles H. KEPNER, Benjamin B. TREGOE, 1965, )
____________________________________
A Systematic Approach to Problem Solving and Decision Making
Charles H. KEPNER
Benjamin B. TREGOE
(Kepner-Tregoe Analytical Methods, root cause analysis, statistical process control (SPC), total quality management(TQM), six sigma)
Introduction
By Perrin Stryker
This book needs no introduction to some 15,000 experienced managers who by 1965 had gone through the training programs developed by Drs. Kepner and Tregoe. It is safe to say that every one of these managers made the same startling discovery that his own private system for handling problems and decisions simply did not work very well and often did not work at all. This is, unfortunately, a discovery that few managers seem to make in the course of their own careers, and they fail to make it largely because the re-education and improvement of their reasoning habits have not been considered
necessary.
Yet the cost unsystematic and irrational thinking by managers is undeniably enormous. If he wants to, any good manager can easily recall from his experience a wide assortment of bungled problems and erroneous decisions. As an executive of a large corporation long honored for its good management once said to me, "The number of undisclosed $10,000 mistakes made in this company every day makes me shudder." However, like others in management, this executive did not think his subordinates could be trained to think more clearly about problems and decisions; he remained inarticulate about his own thought processes and did not seriously question his habits and methods in handling problems and making decisions. This is understandable. It is much easier for a manager to study things like finances, materials, and markets than it is for him to turn his mind upon his own reasoning processes.
Moreover, clear concepts about the reasoning processes used by managers are still scarce. In recent management literature a lot of attention has been paid to the difficulties involved in handling problems and making decisions. (Brief descriptions of some of these studies will be found in bibliography.) But nearly all the theories and systems proposed seem to be either obscurely complex or obscurely thin and superficial, and they commonly confuse the processes of problem analysis with those of decision making. The arrival of computers and data processing appears to have compounded the confusion. On the one hand, managers are urged to apply computers to their business problems, while, on the other hand, they are frequently reminded that such mechanisms and procedures cannot make their decisions for them. The reminder is essential, for, no matter how many computers he uses, the manager himself still has to know how to reason clearly about problems and their possible solutions.
There is, of course, more involved in being rational manager than the ability to think through problems and decisions logically and systematically. For one thing, a manager needs good judgement to make good decisions, and this capacity is itself a compound to experience, values, and innate abilities which may dictate courses of action that are not neccessarily the products of strictly logical reasoning. Setting objectives and policies similarly involves considerations that may not be reasonable from one or another viewpoint. But the capacity to reason systematically is unquestionably a basic neccessity for any manager who hopes to manage well. Successful managers have developed this capacity through experience. But the first chapter of this book shows that even experienced managers are
surprisingly inefficient in the ways they go about handling problems and decisions.
It was this inefficiency that attracted the attention of the authors during the 1950s while they were working for the RAND Corporation in California. For several years up to 1958 Kepner, a social psychologist, and Tregoe, a sociologist, had spent most of their time working with advanced systems of defense for the Air Force, and this "ivory tower" occupation eventually got them thinking about the ways people use information in a highly automatic data-processing system. On their own time they studied the effects of automation in industry, and from there it was only a step to the processes of decision making. They soon saw that a great many business decisions were often wretchedly made and extremely costly. They were convinced that there must be some principles and basic techniques that would improve this managerial performance. From their work at RAND they knew that new directions would have to be followed if they were going to develop the underlying concepts of decision making. So they decided to leave their jobs and undertake this research entirely on their own.
The progress of their research is instructive. First, they reviewed the literature on problem solving and decision making, looking for techniques and concepts or principles that might explain the difference between good and poor decision making. They found bits and pieces, but precious little that they considered useful. Then they examined the internal workings of an organization from policy level to accounting procedures, looking at its complete operation. But these business details did not help in finding concepts that could be used in solving problems. Sitting in Tregoe's garage they spent hour after hour trying first one idea or techniques and then another, but nothing worked out to their satisfaction. Then one day they completely reversed their attack and decided to start with a problem of a company and work backward through the process of solving it, dissecting the thought process involved at each step.
This engineering approach produced entirely different results. They developed a set of ideas based upon what a manager has to do to solve a problem in real life; then, to help make these ideas visible, they developed a simulation of an imaginary company which they call "APEX." The first problem they posed for research was a series of customer complaints about one style of APEX's screen doors. They chose this problem because Tregoe happened to be looking at his own screen door as they sat in his garage, and they figured that some trouble with a door would be an easy thing to visualize.
As they began to think through the detail of this screen door problem, they began to expose some of the common confusions of problem solving; they saw, for instance, that it was no use to advise a manager to "define the problem" before he had even identified which was the most important or urgent problem he had to deal with. Nor was it fruitful for a manager to ask why a problem had occurred before he knew exactly WHAT it was he had to explain. By such pragmatic thinking they followed through the processes that would be involved in reaching a correct solution, and they backed up their findings with more field research. They spent six months asking managers in a variety of businesses about the steps these managers actually took in solving typical problems and making decisions about them.
This research produced many fresh concepts. While the concepts in existing textbooks were based on one discipline
or another, those of Kepner and Tregoe were eclectic and did not fit with even bits and pieces of the established works on
problem solving and decision making. For example, they developed the concept of a problem as a deviation from a standard, and the concept of cause as unplanned, unexpected change, and only later learned that Professor Herbert A. Simon was at about the same time independently arriving at similar concepts in his computer research on problem solving theory. For their analysis of problem solving and decision making, Kepner and Tregoe eventually developed fourteen concepts (summarized in Chapter 3) and then moved on to the process of potential problem analysis which is dealt with in the last chapter of this text. The whole progression took nearly five years to develop to its present stage, and further developments and refinements will undoubtedly follow.
The validity of these concepts and procedures has already been established in the most convincing way, by practical applications on the job. Some of these applications, described in Chapter 9, demonstrated the flexibility that managers have used in adapting the ideas and techniques to special situations. It is also notable that training in the use of the concepts and procedures has become a regular part of management development in some of the leading United States corporations, including General Motors, Ford, Du Pont, General Electric, Honeywell, and IBM.
Managers themselves will probably find the training methods devised by Kepner and Tregoe and their associates to be as impelling as the concepts. The authors present a full explanation of their teaching methods in the appendix, but I would like to emphasize here that this training provides a kind of experience that managers rarely, if ever, get. In this three part training method, the first part--study of the concepts--is familiar enough and parallels the kind of vicarious management experience gotten through lectures, discussions, audio-visual and case study methods. The second part--intensive practice of the concepts and procedures in a simulated business situation--is also not entirely new, though in its sophistication and intensity it differs from in basket techniques, workshops, and role-playing sessions. But the third part of this training--the feedback sessions on actual performance--is not duplicated anywhere else to my knowledge. For these feedback sessions provide the managers with an immediate and detailed critique of their actual performance and show them HOW they went about solving problems and making decisions, WHAT was wrong or might be improved, and HOW they can revise their performance. By way of contrast, managers playing business games with computers are in effect simply guessing the computer's philosophy of business and management; they do not learn how they went wrong and specifically how they can improve their decisions--and, of course, such games teach nothing about the analysis of problems.
The research and training methods developed by Kepner and Tregoe carry certain inevitable and significant implications for anyone interested in the processes of management. Their work clearly shows that problem analysis and decision making are management acts that should be consciously and systematically performed, and if necessary, recorded. The idea that a manager should be conscious of exactly what he is doing while he is managing may not sound revolutionary, but the fact is that such management is seldom found in industry today. The absence of conscious, systematic problem analysis and decision making is not only responsible for inefficiency and waste; it is also responsible, in large part, for the general neglect of two of the most important management functions: the setting of clear objectives, and the setting of clear performance standards for personnel.
Perhaps the most significant implication of the concepts and procedures described in this book is that they anticipate the kind of manager that will be needed in the future. As the authors have pointed out, the continuing increases in technology inevitably mean that managers will know less and less about the skills and knowledge of those they are managing, and will have to depend more and more on their ability to manage the operating techniques of those reporting to them. And in order to manage the way a subordinate handles problems and decisions, a manager has to know how to ask just the right questions. It is precisely with this critical management skill that this book is ultimately concerned. Such skill cannot, of course, be acquired overnight, and one reading of this book will certainly not develop it, but in my judgement there is no better place to start.
CHARLES H. KEPNER, a graduate of Fresno State College, holds an M.S. degree in psychology from University of Oregon and a Ph.D. in social psychology from University of Michigan. He spent several years in contract research in Canada in the fields of mental health, conservation, and medicine,
and was research director for the Royal Commission on Agriculture and Rural Life. Following two years in military training research at the RAND Corporation, he left to join Tregoe and set up their organization to study problem
solving and decision making.
BENJAMIN B. TREGOE, a graduate of Whittier College, holds a Ph.D. degree in sociology from Harvard, and has lectured at University of Southern California and at University of California at Los Angeles. For three years he was field training consultant to several Air Force installations, and was doing systems research at the RAND Corporation before joining Kepner to establish KEPNER-TREGOE AND ASSOCIATES, INC.
1965 by Charles H. KEPNER and Benjamin B. TREGOE
(The Rational Manager : A Systematic Approach to Problem Solving and Decision Making, Charles H. KEPNER, Benjamin B. TREGOE, 1965, )
____________________________________
The Rational Manager
A Systematic Approach to Problem Solving and Decision Making
Charles H. KEPNER
Benjamin B. TREGOE
pp.80-87
Take, for example, the problem that was faced by a manufacturer of an automatic garage door that was electronically powered and motivated by a tiny radio transmitter in the car. The transmitter signal would activate the door, raising it if it were down and lowering it if it were up. The transmitter had an ultrashort range, so that one signal would not raise all the doors in the block. When the doors were first introduced in 1958 they were very successful, but as the fall approached and the weather turned cooler, the fog from the bay began to roll in and complaints also began to roll in. Such as: "Damn thing came right down in the middle of my new covertible--and I didn't touch the transmitter button!" And one woman phoned: "Come and take your door off; it came down and almost killed my husband last night."
The company, facing a major crisis in these complaints, proceeds to specify the problem. The specification showed that the complaints came in only on certain days but not on others, and particularly late in the afternoon and evening. Also, the complaints came from a single area on a part of land extending into the bay. This particular piece of information came out as they plotted the sources of the complaints on a map of the city. It soon became apparent that the complaints came from a strip of land cutting entirely across the residential section of the point. This area of complaints was wider at the ocean end and narrower at the bay end.
Once this specific outline of the problem area had been drawn, the search for distinctions in the specification began. What was distinctive of the days when there were complaints as opposed to the days when there were none? A quick check showed that days of complaints were days of heavy fog. What was distinctive of late afternoon and early evening? This was when the fogs rolled in, if it was going to be foggy. What was distinctive of the narrow coneshaped strip of land from which the complaints came? This strip was directly in line with the end of the longest runway of a Naval air station located on an island out in the bay. Planes crossed that strip of land as they made their final approach.
Thus the distinctions found in the specification were "fog in the air," and "planes across this area on final approach to the longest runway." Now, within, or with respect to, these areas of distinction, what changes took place? During the foggy periods the Naval air station flight controller guided the aircraft in to landings by radar, and he used a special radio frequency. He also used the longest runway exclusively for ground-controlled approaches. Aircraft coming in over the strip of land on foggy days also used the same special radio frequency. A check on this frequency proved it to be the same as the frequency used to operate the garage doors. Every time the Naval flight controller and the incoming pilot would speak to one another during an approach, their radio frequency would run the doors up and down. Once this was recognized by the door manufacturer, the door mechanisms were changed to a new frequency, and the trouble disappeared. After the fact, everyone recalled that all the doors had opened or shut about the same time, that there was always a plane overhead, and that it was always foggy at the time. But they had not recognized these things at the outset. Their experience didn't help them see the pertinent facts.
p.82
FIGURE 15 This diagram shows the distinction that led to the cause of a mysterious opening of garage doors in one locality. As the clues to cause, distinctions are a major tool of problem solving. If a cause has an effect one place but not another, there MUST be something distinctive of that one place to make this happen.
Similarly, a manager's technical know-how and experience cannot be expected to guarantee a precise specification of a problem in his area. Technical experience, in fact, can easily blind managers to the very facts that would lead to solution. This was demonstrated in a plant of a large paper manufacturer where there is a pulping plant and a paper machine in line, with softwood logs being chipped and boiled with certain chemicals to the right consistency, and then fed as pulp into the papermaking machine. The plant was working well when suddenly it was discovered that small pieces of wood were coming through in the dried, finished sheets of paper. It was immediately assumed that something was wrong in the pulping process, that one of the huge stainless-steel screens had broken. Whereupon some $70,000 worth of new equipment was ordered to correct the problem.
However, one man was not blinded by his papermaking experience. He did not take the "obvious" explanation. Instead, he closely examined the troublesome wood pieces in the paper and found that these were not softwood chips but hardwood splinters; moreover, they had never been cooked or chemically treated. Then he spotted a hardwood pipeline used to transfer the pulp to the papermaking machine. He told his colleagues that the lining of this pipe must be breaking up on the inside and letting hardwood splinters get into the pulp mixture. His colleagues thought he was crazy because they had never heard of a hardwood pipe breaking up on the inside. But his explanation was finally checked out and found to be so, proving that the problem had nothing at all to do with the pulping equipment. This might have been determined at the outset if the problem had originally been precisely specified as "uncooked hardwood splinters in finished paper," instead of simply "piece of wood in the paper."
Precise time can be critically important in a specification. For example, some time after one cocoa manufacturer introduced a new combination metal-and-cardbroad container, loud complaints began coming in from house wives. They protested that when they opened the new cocoa cans they got a very strong disinfectant smell out of the can. However, the company found that other housewives using the same brand of cocoa in the same new cans had no complaints about odor at all. The trouble was only traced down after time was specified as one of the characteristics of the deviation. When the company followed up on the lot numbers of the cocoa cans that were bringing in the complaints, it turned out that the cocoa that smelled of chlorine disinfectant was distinctive in that it had been on the dealers' shelves for "three months or more." And cocoa packed in the same kind of containers and taken from the same lots, which had been on the shelves for less than three months did not smell of chlorine.
This distinction in shelf time spotted the change that caused the problem: the container manufacturer had begun using a new chlorine disinfectant to sterilize the wood pulp before forming the container, as required by law. This chlorine residue in the cardboard was liberated over a period of three months or more into the air space of the cocoa container, producing the disinfectant smell. It was an expensive change: the cocoa manufacturer collected $800,000 in damanges against the container manufacturer.
The critical information that must go into a specification is not always easy to recognize. A good deal of questioning is often necessary to dig it out, particularly with respect to distinctions and changes. Such was the case with a major engineering and electronics company that developed secretary trouble after one of its reorganizations. The girls became dissatisfied, then unhappy, then furious. They complained that nothing were no good, that their desks were not level, that the air conditioning was noisy, and that management decisions were getting worse and worse. Attempts to specify the exact nature of the complaints led to much emotional catharsis but little hard data, as sometimes happens in situations like this. However, specification of "What Object"--in this case "who," i.e., the individual girls involved,--was somewhat easier. These girls turned out to be only about a fifth of the secretaries employed. Specifying "Where Observed" located all the dissidents in an older buildings--but not all of the girls in this older building were complaining.
Then it took a lot of probing to recognize what was distinctive of these complaining girls in the older building. The distinction was that all of them had been moved during the recent reorganization. Shortly thereafter the relevant changes were discovered: in each case, the girl had been moved from a new building to the older building, or from a larger to a smaller office within the older building. These changes were the cause of the problem: the girls were resenting the loss of status implied in the moves. Nothing was wrong with the typewriters, desks, air conditioning, or management's decisions, with the possible exception of the lack of foresight shown when considering the moves. But as long as the problem was analyzed in terms of the complaints themselves, little progress was made toward finding the cause. As soon as the girls who were complaining were treated as the IS, and the noncomplainers as the IS NOT, it was possible to move ahead. Even the, the critical distinction was not recognized until after much probing. But then it became obvious: "Of course, why didn't I think of that before?"
These cases, and many more like them, repeatedly demonstrate how necessary it is for the manager to use both exact and inclusive observation in developing the specification. Since every problem differs somewhat from every other problem, the facts in a specification must represent the uniqueness of the deviation precisely, or the clues to the cause may be missed. As the specification illustrated on page 77 shows, the question are all about what, where, when, and how big. The question "why?" is never asked, for this can only be answered when the cause is known. Asking "why?" is an invitation to speculate loosely about causes. If such speculation is allowed to enter into a specification, the tendency will be to develop facts that "prove" a cause which the manager suspects or has a hunch about. This does not mean that he need throw his hunches away, but he should set them aside until he has precisely specified the problem and analyzed it for the distinctions and changes that will lead him to possible causes. Then he may test his hunches as to changes that have occurred, along with the possible causes he had derived from this analysis.
As a practical matter, it does not take much time to set out the elements of a precise specification. He asks WHAT is wrong and WHAT OBJECT is affected; WHERE on the object the deviation occurs and WHERE objects with such deviation are observed; WHEN the deviation appears on the object and WHEN objects with the deviation are observed; and HOW BIG the deviation is and HOW MANY objects with deviations have been observed. A manager can then go through the facts about a deviation rather swiftly as he concentrates on separating what the deviation IS from what IS NOT the deviation but is closely related to it.
Once he has described and drawn an exact line around the deviation, he can proceed to analyze the specification for clues to the problem's cause. As already indicated, this process of analysis includes two stages: (1) a search for those characteristics distinctive of the IS but not of the IS NOT in the specification, and (2) a search for changes within such areas of distinction. These clue-seeking procedures will be spelled out in the following chapters.
p.96
... Problems are often made more difficult by the element of time. Time distinctions are therefore often critical in the solution of a problem. Time distinctions are therefore often critical in the solution of a problem. Thus the effects produced by a change may not be noticed immediately, and this time lag between change and observed effect may cover up the clues needed to relate cause and problem. Or it may be that time is required for the change to produce an effect, ...
p.97
... Without complete knowledge of the problem you sometimes cannot recognize the answer when you have it.
p.98
... Often one of the smallest and seemingly most inconsequential details will be the one that identifies a distinction that points to the change that caused the problem.
p.99
... There are simply too many possibilities. A distinction must be found to narrow the search for change down to within reasonable limits.
p.99
([... missiles testing for assembly, manufacturing and quality assurance issues, in one case study the problem was with how the wiring inside the missile assembly was tied-down (by the way, wiring problems was not detected in quality assurance), in second case, it was the fin of the missile, the fin came loose at high-altitude because of the cold and metal contraction, causing the fin to wobble during high-speed flight, neither problems was obvious because the sometimes the fin was secured enough and the missile was not tested at high altitudes each and every time for enough length of time to cause the metal to contract to cause the fin to wobble during flight, and for the wiring problem, that was a production issue and if the right guy was tying down and doing the wiring, there would be no problem with that missiles batch. ])
© 1965 by Charles H. KEPNER and Benjamin B. TREGOE
(The Rational Manager : A Systematic Approach to Problem Solving and Decision Making, Charles H. KEPNER, Benjamin B. TREGOE, © 1965, pp.80-87, p.96, p.97, p.98, p.99)
A Systematic Approach to Problem Solving and Decision Making
Charles H. KEPNER
Benjamin B. TREGOE
pp.80-87
Take, for example, the problem that was faced by a manufacturer of an automatic garage door that was electronically powered and motivated by a tiny radio transmitter in the car. The transmitter signal would activate the door, raising it if it were down and lowering it if it were up. The transmitter had an ultrashort range, so that one signal would not raise all the doors in the block. When the doors were first introduced in 1958 they were very successful, but as the fall approached and the weather turned cooler, the fog from the bay began to roll in and complaints also began to roll in. Such as: "Damn thing came right down in the middle of my new covertible--and I didn't touch the transmitter button!" And one woman phoned: "Come and take your door off; it came down and almost killed my husband last night."
The company, facing a major crisis in these complaints, proceeds to specify the problem. The specification showed that the complaints came in only on certain days but not on others, and particularly late in the afternoon and evening. Also, the complaints came from a single area on a part of land extending into the bay. This particular piece of information came out as they plotted the sources of the complaints on a map of the city. It soon became apparent that the complaints came from a strip of land cutting entirely across the residential section of the point. This area of complaints was wider at the ocean end and narrower at the bay end.
Once this specific outline of the problem area had been drawn, the search for distinctions in the specification began. What was distinctive of the days when there were complaints as opposed to the days when there were none? A quick check showed that days of complaints were days of heavy fog. What was distinctive of late afternoon and early evening? This was when the fogs rolled in, if it was going to be foggy. What was distinctive of the narrow coneshaped strip of land from which the complaints came? This strip was directly in line with the end of the longest runway of a Naval air station located on an island out in the bay. Planes crossed that strip of land as they made their final approach.
Thus the distinctions found in the specification were "fog in the air," and "planes across this area on final approach to the longest runway." Now, within, or with respect to, these areas of distinction, what changes took place? During the foggy periods the Naval air station flight controller guided the aircraft in to landings by radar, and he used a special radio frequency. He also used the longest runway exclusively for ground-controlled approaches. Aircraft coming in over the strip of land on foggy days also used the same special radio frequency. A check on this frequency proved it to be the same as the frequency used to operate the garage doors. Every time the Naval flight controller and the incoming pilot would speak to one another during an approach, their radio frequency would run the doors up and down. Once this was recognized by the door manufacturer, the door mechanisms were changed to a new frequency, and the trouble disappeared. After the fact, everyone recalled that all the doors had opened or shut about the same time, that there was always a plane overhead, and that it was always foggy at the time. But they had not recognized these things at the outset. Their experience didn't help them see the pertinent facts.
p.82
FIGURE 15 This diagram shows the distinction that led to the cause of a mysterious opening of garage doors in one locality. As the clues to cause, distinctions are a major tool of problem solving. If a cause has an effect one place but not another, there MUST be something distinctive of that one place to make this happen.
Similarly, a manager's technical know-how and experience cannot be expected to guarantee a precise specification of a problem in his area. Technical experience, in fact, can easily blind managers to the very facts that would lead to solution. This was demonstrated in a plant of a large paper manufacturer where there is a pulping plant and a paper machine in line, with softwood logs being chipped and boiled with certain chemicals to the right consistency, and then fed as pulp into the papermaking machine. The plant was working well when suddenly it was discovered that small pieces of wood were coming through in the dried, finished sheets of paper. It was immediately assumed that something was wrong in the pulping process, that one of the huge stainless-steel screens had broken. Whereupon some $70,000 worth of new equipment was ordered to correct the problem.
However, one man was not blinded by his papermaking experience. He did not take the "obvious" explanation. Instead, he closely examined the troublesome wood pieces in the paper and found that these were not softwood chips but hardwood splinters; moreover, they had never been cooked or chemically treated. Then he spotted a hardwood pipeline used to transfer the pulp to the papermaking machine. He told his colleagues that the lining of this pipe must be breaking up on the inside and letting hardwood splinters get into the pulp mixture. His colleagues thought he was crazy because they had never heard of a hardwood pipe breaking up on the inside. But his explanation was finally checked out and found to be so, proving that the problem had nothing at all to do with the pulping equipment. This might have been determined at the outset if the problem had originally been precisely specified as "uncooked hardwood splinters in finished paper," instead of simply "piece of wood in the paper."
Precise time can be critically important in a specification. For example, some time after one cocoa manufacturer introduced a new combination metal-and-cardbroad container, loud complaints began coming in from house wives. They protested that when they opened the new cocoa cans they got a very strong disinfectant smell out of the can. However, the company found that other housewives using the same brand of cocoa in the same new cans had no complaints about odor at all. The trouble was only traced down after time was specified as one of the characteristics of the deviation. When the company followed up on the lot numbers of the cocoa cans that were bringing in the complaints, it turned out that the cocoa that smelled of chlorine disinfectant was distinctive in that it had been on the dealers' shelves for "three months or more." And cocoa packed in the same kind of containers and taken from the same lots, which had been on the shelves for less than three months did not smell of chlorine.
This distinction in shelf time spotted the change that caused the problem: the container manufacturer had begun using a new chlorine disinfectant to sterilize the wood pulp before forming the container, as required by law. This chlorine residue in the cardboard was liberated over a period of three months or more into the air space of the cocoa container, producing the disinfectant smell. It was an expensive change: the cocoa manufacturer collected $800,000 in damanges against the container manufacturer.
The critical information that must go into a specification is not always easy to recognize. A good deal of questioning is often necessary to dig it out, particularly with respect to distinctions and changes. Such was the case with a major engineering and electronics company that developed secretary trouble after one of its reorganizations. The girls became dissatisfied, then unhappy, then furious. They complained that nothing were no good, that their desks were not level, that the air conditioning was noisy, and that management decisions were getting worse and worse. Attempts to specify the exact nature of the complaints led to much emotional catharsis but little hard data, as sometimes happens in situations like this. However, specification of "What Object"--in this case "who," i.e., the individual girls involved,--was somewhat easier. These girls turned out to be only about a fifth of the secretaries employed. Specifying "Where Observed" located all the dissidents in an older buildings--but not all of the girls in this older building were complaining.
Then it took a lot of probing to recognize what was distinctive of these complaining girls in the older building. The distinction was that all of them had been moved during the recent reorganization. Shortly thereafter the relevant changes were discovered: in each case, the girl had been moved from a new building to the older building, or from a larger to a smaller office within the older building. These changes were the cause of the problem: the girls were resenting the loss of status implied in the moves. Nothing was wrong with the typewriters, desks, air conditioning, or management's decisions, with the possible exception of the lack of foresight shown when considering the moves. But as long as the problem was analyzed in terms of the complaints themselves, little progress was made toward finding the cause. As soon as the girls who were complaining were treated as the IS, and the noncomplainers as the IS NOT, it was possible to move ahead. Even the, the critical distinction was not recognized until after much probing. But then it became obvious: "Of course, why didn't I think of that before?"
These cases, and many more like them, repeatedly demonstrate how necessary it is for the manager to use both exact and inclusive observation in developing the specification. Since every problem differs somewhat from every other problem, the facts in a specification must represent the uniqueness of the deviation precisely, or the clues to the cause may be missed. As the specification illustrated on page 77 shows, the question are all about what, where, when, and how big. The question "why?" is never asked, for this can only be answered when the cause is known. Asking "why?" is an invitation to speculate loosely about causes. If such speculation is allowed to enter into a specification, the tendency will be to develop facts that "prove" a cause which the manager suspects or has a hunch about. This does not mean that he need throw his hunches away, but he should set them aside until he has precisely specified the problem and analyzed it for the distinctions and changes that will lead him to possible causes. Then he may test his hunches as to changes that have occurred, along with the possible causes he had derived from this analysis.
As a practical matter, it does not take much time to set out the elements of a precise specification. He asks WHAT is wrong and WHAT OBJECT is affected; WHERE on the object the deviation occurs and WHERE objects with such deviation are observed; WHEN the deviation appears on the object and WHEN objects with the deviation are observed; and HOW BIG the deviation is and HOW MANY objects with deviations have been observed. A manager can then go through the facts about a deviation rather swiftly as he concentrates on separating what the deviation IS from what IS NOT the deviation but is closely related to it.
Once he has described and drawn an exact line around the deviation, he can proceed to analyze the specification for clues to the problem's cause. As already indicated, this process of analysis includes two stages: (1) a search for those characteristics distinctive of the IS but not of the IS NOT in the specification, and (2) a search for changes within such areas of distinction. These clue-seeking procedures will be spelled out in the following chapters.
p.96
... Problems are often made more difficult by the element of time. Time distinctions are therefore often critical in the solution of a problem. Time distinctions are therefore often critical in the solution of a problem. Thus the effects produced by a change may not be noticed immediately, and this time lag between change and observed effect may cover up the clues needed to relate cause and problem. Or it may be that time is required for the change to produce an effect, ...
p.97
... Without complete knowledge of the problem you sometimes cannot recognize the answer when you have it.
p.98
... Often one of the smallest and seemingly most inconsequential details will be the one that identifies a distinction that points to the change that caused the problem.
p.99
... There are simply too many possibilities. A distinction must be found to narrow the search for change down to within reasonable limits.
p.99
([... missiles testing for assembly, manufacturing and quality assurance issues, in one case study the problem was with how the wiring inside the missile assembly was tied-down (by the way, wiring problems was not detected in quality assurance), in second case, it was the fin of the missile, the fin came loose at high-altitude because of the cold and metal contraction, causing the fin to wobble during high-speed flight, neither problems was obvious because the sometimes the fin was secured enough and the missile was not tested at high altitudes each and every time for enough length of time to cause the metal to contract to cause the fin to wobble during flight, and for the wiring problem, that was a production issue and if the right guy was tying down and doing the wiring, there would be no problem with that missiles batch. ])
© 1965 by Charles H. KEPNER and Benjamin B. TREGOE
(The Rational Manager : A Systematic Approach to Problem Solving and Decision Making, Charles H. KEPNER, Benjamin B. TREGOE, © 1965, pp.80-87, p.96, p.97, p.98, p.99)
____________________________________
