Problem Solving for Process Operators and Specialists

By Joseph M. Bonem

This book provides methods to train process operators to solve challenging problems. The book is split into two parts. The first part consists of two parts; first developing a daily monitoring system and second providing a structured 5 step problem solving approach that combines cause and effect problem solving thinking with the formulation of theoretically correct hypotheses. The 5 step approach emphasizes the classical problem solving approach (defining the sequence of events) with the addition of the steps of formulating a theoretically correct working hypothesis, providing a means to test the hypothesis, and providing a foolproof means to eliminate the problem. The initial part of the book focuses on defining the problem that must be solved and obtaining the location, time and quantity based specifications of the problem. This part of the book also presents techniques to find and define problems at an early point before they progress to the critical level.

The second part of the book deals with the utilization of fundamental chemical engineering skills to develop a technically correct working hypothesis that is the key to successful problem solving. The primary emphasis is on simple pragmatic calculation techniques that are theoretically correct. It is believed that any operator can perform these calculations if he is provided the correct prototype. Throughout the book, the theory behind each pragmatic calculation technique is explained in understandable terms prior to presenting the author's approach. These techniques have been developed by the author in 50+ years of industrial experience. The book includes many sample problems and examples of real world problem solving. Using these techniques, theoretically correct working hypotheses can be developed in an expedient fashion.

• Language: English
• Publisher: Wiley
• Release date: Apr 18, 2011
• ISBN: 9780470934623

Book preview

INITIAL CONSIDERATIONS

1.1 INTRODUCTION

Problem solving is found throughout all activities of daily life. Problem solving tends to take place in two mind modes. There is the intuitive or instinctive reactionary mode, which has also been called gut feel. Then there is the methodical reasoning approach, which is usually based on theoretical considerations and calculations.

Both of these approaches have a place in real world problem-solving activities. The intuitive reactionary person will respond much faster to a problem. The response is usually based on experience. That is, he has seen the same thing before or something very similar and remembers what the problem solution was. However, if what is occurring is a new problem or is somewhat different, his approach may lead to an incorrect problem solution. The methodical reasoning person will not be able to react to problems quickly, but will usually obtain the correct problem solution for complicated problems much faster than the intuitive reactionary person, who must develop and perhaps discard several gut feel solutions.

Here is an example of how two people with these different mind-sets can react. On a golf course, the cry of Fore will elicit different responses. The person responding based on intuition or instinct will immediately crouch and cover his head. This will reduce the probably that the errant golf ball hits a sensitive body part. The person responding based on methodical reasoning will begin to assess where the cry came from and where the ball might be coming from, and then reach a conclusion as to where it might land. Obviously, in this case, reacting based on intuition or instinct is a far superior mode of operating. There are many more examples from the sports world where reacting in an intuitive fashion yields far superior results than reacting in a methodical reasoning manner. However, essentially all of these examples will be experience-based. People who are reacting successfully in an intuitive mode know what to do because they have experienced the same or very similar situations.

Similar things happen in industrial problem solving. Experienced people such as engineers or operators react instinctively because they have experienced similar events. These operators or engineers do an excellent job of handling emergency situations or making decisions during a startup. As a rule, the person who tends to respond based on methodical reasoning and calculations can rarely react fast enough to be of assistance in an emergency or if quick action is required in a startup situation. The exception to this rule is the engineer who has designed the plant and has gone through calculations to understand what will happen in an emergency or startup. In effect, he has gained the experience through calculations as opposed to actual experience.

The experience necessary to conduct problem solving in the real world does not always exist. In addition, while the need for quick response when solving industrial problems is real, there is not always an emergency or crisis that requires immediate action. Thus the methodical reasoning approach is often the desirable mode of operating. The three components of this methodical reasoning approach are:

1. A systematic, step-by-step procedure. This includes the three essential problem-solving skills (Daily Monitoring System, Disciplined Problem-Solving Approach, and Determining Optimum Technical Depth).

2. A good understanding of how the equipment involved works.

3. A good understanding of the specific technology involved.

Before discussing problem solving in industrial facilities, two examples from everyday life are discussed. It often aids learning to discuss things that are outside the scope of the original thrust of the teaching. The two examples from everyday life discussed below will be helpful in understanding the difference between intuitive problem solving and that based on methodical reasoning.

1.2 AN ELECTRICAL PROBLEM

While trimming bushes with an electric hedge trimmer, a laborer accidentally cut the extension cord being used to power the trimmer. He had been using an electrical outlet in a pump house located approximately 70 ft from the main house. The only other use for 110 volt electricity in the pump house was for a small clock associated with the water softener. The laborer found another extension cord and replaced the severed cord. However, when he plugged it in and tried to turn on the hedge trimmer, it did not have any power. He then had to report the incident to the homeowner. The homeowner checked the panel-mounted circuit breakers. None of them appeared to be tripped. To be sure, he turned off the appropriate circuit breaker and reset it. However, power was still not restored to the outlet in the pump house. To make sure that the replacement extension cord was not the problem, the homeowner plugged another appliance into the electrical outlet in the pump house. It did not work either. The homeowner then concluded that the electric outlet had been blown out when the cord was cut. He replaced the electric outlet. However, this still did not provide power to the equipment. When the homeowner rechecked the circuit breaker, he noticed that a ground fault interrupter (GFI) in a bathroom in the main house was tripped. Resetting this GFI solved the problem. Ground fault interrupters are designed to protect from electrical shock by interrupting a household circuit when there is a difference in the currents in the hot and neutral wires. Such a difference indicates that an abnormal diversion of current from the hot wire is occurring. Such a current might be flowing in the ground wire, such as a leakage current from a motor or from capacitors. More importantly, that current diversion may be occurring because a person has come into contact with the hot wire and is being shocked.

While the homeowner believed that in this particular house every GFI protected a single outlet, it is not unheard-of to protect more than a single outlet with a GFI. It seemed surprising that the GFI in a bathroom also protected an outlet in the pump house 70 ft away. The homeowner then recalled that at some point in the past, he had noticed that the small clock in the pump house was about 2 hours slow. This clock was always very reliable. In retrospect, he remembered that at about the same time that the clock lost 2 hours, this particular GFI in the bathroom had tripped during a lightning storm and had not been reset for a few hours. Thus it became obvious that the accidental cutting of the extension cord had caused the GFI to trip rather than tripping the circuit breaker or blowing out the electrical outlet. The failure to correctly identify the problem cost the homeowner a small amount of money for the electrical plug and a significant amount of time to go to town to purchase the plug and then install it.

Note that the homeowner’s intuitive conclusions were all valid possibilities. That is, the circuit breaker could well have tripped, the replacement extension cord could have had an electrical break in it, or the electrical outlet could have failed when the original extension cord was cut. His problem solving just did not go into enough detail to solve the problem quickly. Several lessons can be learned from this example. While it seemed to be a simple problem that could be easily solved based on the homeowner’s experience, the intuitive approach did not work. A more systematic approach based on methodical reasoning might have improved results as follows:

Consideration would have been given to the possibility that GFIs can protect more than one electrical outlet. The distance between the GFI and the electrical outlet would not be a consideration. The homeowner did not fully understand the technology.

A voltmeter would have been used to check that power was available coming to the electrical outlet. If power was not available coming to the outlet, the blown plug hypothesis would be invalid. A systematic approach was not used.

In addition, a systematic approach would have raised the question of whether the clock losing 2 hours could be related to the lack of power at the electrical plug.

1.3 A COFFEEMAKER PROBLEM

A man experienced problems with a coffeemaker when it overflowed about half of the time when he made either a flavored or decaffeinated coffee. The coffee and coffee grounds would overflow the top of the basket container and spill all over the counter. The coffee maker performed flawlessly when regular coffee was used. A sketch of the coffee maker is shown in Figure 1-1. When the coffeemaker is started, water is heated and the resulting steam provides a lifting mechanism to carry the mixture of water, steam, and entrained air into the basket where the ground coffee is located. The hot water flows through the coffee and into the carafe. The coffeemaker is fitted with a cutoff valve that causes the flow out of the basket to stop if anyone pulled the carafe out while coffee is still being made.

Figure 1-1 Coffeemaker schematic.

c01f001

The man, a graduate engineer, attempted to determine what was wrong. He examined the problem by first convincing himself that he was following directions when it came to making the coffee. He then carefully examined the equipment, especially the cutoff valve. He concluded that somehow the cutoff valve was restricting the liquid flow whenever decaffeinated or flavored coffee was being made. That is, the incoming flow of hot water and steam was greater than the flow out of the valve. This would cause the level in the container to build up and run over. The problem solution seemed relatively simple. He removed the valve and made a sign that read, Do not remove carafe until coffee is finished brewing. He felt a surge of pride in not only solving the problem, but that he prevented a future problem by providing instructions to prevent someone from pulling out the carafe. The next time that one of the suspect coffees was made, the container did not overflow. He then announced that the problem was solved.

Unfortunately, the glow of successful problem solving did not last long. The next time that flavored coffee was made, the problem recurred; that is, the coffee and grounds flowed over the top of the basket container. The engineer then began a more detailed investigation of the problem, including understanding the technology for making flavored and decaffeinated coffee. He discovered that when decaffeinated coffee was produced at the coffee supplier, a surface active material was utilized. This surface active material was mixed with the coffee to extract the caffeine. Materials that are surface active have the capability to thoroughly contact the coffee solid so that caffeine is removed from not only the surface, but the deep pores. The surface active material also reduces the surface tension of water, which creates a system that can easily foam.

The engineer then extrapolated from this knowledge and theorized that when flavored coffee was made, a surface active material was used to evenly distribute the flavor to the coffee. Once that he understood the difference in the coffee making processes, he theorized that residual amounts of the surface active material being left on the coffee reduced the surface tension of the hot water and coffee and caused it to foam up in the container and out over the sides onto the counter.

Since the amount of residual surface active material would vary slightly from batch to batch, it was theorized that only the batches of either flavored or decaffeinated coffee that contained greater than a critical level would cause an overflow. After studying this theory, the engineer decided that the problem solution would be to obtain a coffeemaker that had a basket container with a different design. The problematic coffeemaker had a small cylindrical basket. A new coffeemaker with a large conical design basket was purchased. The comparison of the two baskets is shown in Figure 1-2. It was theorized that the large conical design would provide a reduced upward velocity of the foaming material and this would allow release of the vapor trapped in the foam. The purchase of this coffeemaker eliminated the problem completely.

Figure 1-2 Basket comparisons.

c01f002

Several lessons can be learned from this problem-solving exercise. The intuitive hunch that coffee was not flowing through the valve as fast as hot water was coming into the basket made logical sense. However, no logical explanation was provided for why this only happened with flavored or decaffeinated coffee. Any theory that includes the phrase for some reason is suspect and is an indication of an incomplete problem analysis. A portion of an incomplete problem analysis is almost always logical. However, it is imperative that the entire analysis be logical. Another error was that in formulating the hypothesis, the engineer assumed that only liquid water and solid coffee existed in the container. He overlooked the fact that steam vapors and entrained air were always carried into the container with the hot water. The presence of steam and air would provide a mechanism for creating a frothy mixture. The example also illustrates the need for the following:

A systematic approach, as will be described later in this book, would have eliminated the incomplete hypothesis that suggested the outlet valve was a restriction on only certain grades of coffee.

A sound understanding of how the equipment works: If the engineer had understood how the coffee maker worked, he would not have assumed that only a liquid was present along with the coffee in the container. He would have recognized that both steam and air were carried over into the container along with the hot water.

A sound understanding of the technology involved: The fact that decaffeinated and flavored coffee performed differently than did regular coffee should have been an indication to the engineer that he needed to examine the difference in the coffee-making technology.

These relatively simple examples of how successful problem solving requires a more detailed analysis than simple logic and/or intuition are meant to set the stage for the next chapter, which deals with limitations to industrial problem solving. While industrial problems are almost always more complicated than those described in this section, they require the same problem-solving approaches.

1.4 CLASSIFICATION OF INDUSTRIAL PROBLEMS

It will be of value to classify problems into four categories. This will help determine what kind of effort is required to solve the problem. These categories are as follows:

1. Problems that can be solved based strictly on experience and/or instinct. These are the problems that are typically solved during a startup and/or upset condition. In these situations, there is minimal time for analysis. Experience and instinct are the only way to solve problems in the time available. As you would imagine, the best problem solvers in this situation are those with experience with the particular problem being encountered.

2. Equipment problems that can be solved by application of "first principles. The definition of first principles" is knowledge that has been summarized as a series of mathematical relationships or expressions. An example of this might be a compressor that is not performing as desired. A study of the head curve (a relationship between flow and pressure head expressed in feet) might reveal that the gas flow and/or molecular weight of the gas have changed so that the anticipated pressures can no longer be achieved.

3. Process technology problems that can be solved by application of "first principles. These are process technology problems that can be solved because there are known relationships available. These relationships are often provided in licensing packages or operating instructions. For example, a reactor productivity problem related to impurities in the feed could be solved by using a simplified productivity model. The base line for this productivity model will be based on a licensing package or experimental results. The deviation from the base line would be based on laboratory results and a dynamic model. The simple dynamic model would provide a relationship between time and reactor productivity. These models based on the process technology could be used to show that the loss of productivity correlates with spikes" in a feed contaminant.

4. Process technology problems that cannot be solved by first principles. These might be problems which do not have any reliable solution theory or for which no reliable theory can be developed that relates to the cause of the problem. In other words, theoretically correct first principles do not exist. These are usually very complicated problems involving highly qualitative and subjective variables such as reactor fouling or product attributes such as color, haze, turbidity, or roughness. These subjective variables are often controlled by several independent variables, some of which are well hidden. Some of these controlling variables may be present below the level of analytical detectability. The analysis of such problems is beyond the scope of this book. Fortunately, this classification amounts to only a very small percentage of industrial problems.

The majority of problems in the process industry fall into category 2 or 3. These are the types of problems that the techniques discussed in this book were developed to solve. With an experienced workforce, some of the problems in category 2 or 3 can be solved based on previous history or intuition. However, this experienced workforce is rapidly becoming history. In western countries, the experience level is decreasing as the baby boomers reach retirement age. In developing countries, the workforce is just beginning to build experience. Thus there will be an increasing emphasis on using quantitative methods of problem solving. In addition, cost pressures are driving organizations to reduce the number of graduate engineers in operating plants and to use process operators, specialists, or mechanics as the primary problem solvers.

The problem solver will find it helpful to consider which of the above categories best describes a new problem he is trying to solve. This will aid him in determining what kind of effort is required to solve the problem.

2

LIMITATIONS TO PLANT PROBLEM SOLVING

2.1 INTRODUCTION

While later chapters will consider the structured approach to problem solving, any book dealing with plant problem solving will touch on the question, Is problem solving really part of my job description? The paradigm of this book is twofold.

It is based on the concept that all people working in industrial plants have problem solving as part of their job description whether it is written or not.

To a great extent, the modern process industry has placed operators and process specialists into roles of solving problems. For this problem solving to be done efficiently, they must use some engineering knowledge and calculations. Thus this book discusses engineering problem solving, meaning problem solving that can be done by engineers or operators using engineering calculations.

The first step in developing an effective problem-solving approach is to have the correct mind-set. Some operators and specialists believe that their job is only to turn valves or make educated guesses. At the other end of the spectrum some engineers raise the question, Is problem solving really engineering? Often, engineers may conclude that problem solving is not truly engineering because of the following:

Engineering is defined in such narrow terms that only design work appears to be engineering.

Intuition and gut feel have replaced thorough analysis as a preferred tool for problem solving.

Considerations of optimum technical depth are not well understood.

If one defines engineering, as dictionaries do, as The science of making practical application of knowledge in any field, we must conclude that problem solving is truly engineering. In addition, this definition of engineering also fits an operator with engineering training who is working not just to turn valves, but to solve problems.

It is also important to understand why a course in engineering problem solving is of value. In an example of a typical industrial problem, a customer is unhappy with the appearance of the plastic pellets being received from his supplier. Specifically, the pellets have visual discontinuities similar in appearance to gas bubbles. The customer describes these as voids. If a particle has more than a single void, it is described as a multi-void particle. A simplified statement of the problem is shown in Figure 2-1. As shown in the figure, the process in which the pellets are manufactured consists of two parts, polymerization and extrusion. In the polymerization section, propylene is polymerized to polypropylene particles (700 microns in diameter) using a catalyst. In the extrusion area, these particles are melted, extruded, and formed into cylinders approximately 1/16 by 1/8 in. A strong correlation was developed between the pellet appearance (fraction of pellets with multi-voids) and the polymerization production rate. The problem solver recommended that the production rate be reduced to solve the multi-void problem. This solution to the problem (reducing production rate) is, at best, only a short-range solution. This solution cannot be considered a lasting solution because of the following:

The basic cause of the voids was not considered.

The solution required a severe economic penalty (it might have solved one problem, but it created another one). In most process industries, the limited profits are made at production rates above 75 or 80% of capacity.

Since the basic cause of the voids was not discovered, the problem will likely recur even at the reduced production rates.

Figure 2-1 An example of improper problem solving.

c02f001

2.2 LIMITATIONS TO PROBLEM SOLVING

The previous example is typical of much of the improper problem solving that occurs in many industries in today’s hectic, fast-paced society. It also illustrates why a course in engineering problem solving is of value. There are ten primary limitations to problem solving in today’s process plants. They are described as follows.

1. Modern-day processing plants are large and complex. For example, a relatively simple process such as propylene purification has evolved from fractionation followed by a drying process to remove water to a process incorporating heat pump fractionation and more complicated conversion steps to remove impurities to the part per billion (ppb) level. In addition, plant sizes have increased significantly. Thus, there is even more emphasis on solving problems quickly and correctly.

2. The problem is usually more complicated than first described. Typical initial problem descriptions might consist of such statements as, It won’t work as designed, or, It won’t work unless you modify it to … If either of these problem descriptions is accepted exactly as stated, the problem solver is doomed to failure. In order to practice true engineering problem solving, the problem solver must use a disciplined approach that involves writing out an accurate description of the problem that does not include a problem solution. This is necessary to avoid ignoring data and jumping to conclusions.

3. Conflicting data will always be present, and can take many forms. Some examples are that the verbal descriptions eyewitnesses give can disagree; laboratory data may be in disagreement with physical factors, instrumentation, or even other laboratory data; and/or instrumentation/computer data may be in conflict with other sources of data.

4. Modern day plants have a great deal of variable interaction. This results in difficulty in isolating the real problem affecting either independent variables or strong correlations between dependent variables. While a strong correlation between dependent variables may be of interest, it rarely results in the solution to problems. An independent variable can be changed or set by an operator or by operating directives. Dependent variables are those that are changed by the reaction of the process. In the plastic pellet example given earlier, the independent variable is the production rate and the dependent variable is the fraction of particles with voids.

5. Besides a high degree of variable interaction, there is also a high degree of interaction between various engineering disciplines. Thus, what appears to be an obvious mechanical engineering problem often has its true roots in chemistry and/or chemical engineering. The converse is also true. This confusing scenario often leaves the process operator caught in the middle, not knowing which course of action to pursue.

6. System dynamics involve long holdup times. In the modern day process, there is usually an incentive to push the process to higher efficiency or higher purity. This usually leads to longer residence times in equipment. Problem solving with long residence time equipment requires the use of a dynamic model. Unfortunately, when faced with the need for a dynamic model, the problem solver will often take one of two unsatisfactory approaches. He will give up on the basics and say, It’s too complicated. Since the dynamic model is truly required to solve the problem, the problem solver must now take an approach that can be characterized as guess work. The other extreme is that he will begin the development of an elaborate, technically correct model that will probably not be finished in time to be of any assistance. Both of these approaches overlook the fact that there are ways to build simple, technically correct dynamic models. These simple models will contain assumptions, however, these assumptions will still provide a model with sufficient accuracy to solve industrial problems.

7. Engineering principles are often inadequately applied by operators as well as engineers. In today’s industrial environment, pressures to perform at a minimum cost and manpower commitment often encourage shooting from the hip as a problem-solving technique. This may be completely appropriate in some limited situations. However, the purpose of this book is to address the chronic problem that is only wounded by the shoot from the hip technique. The modern chemical engineering curriculum, while providing an excellent theoretical foundation, often fails to adequately stress the application of fundamentals. For example, Bernoulli’s theorem can be used to explain inaccurate values given by the poorly designed level instrument shown in Figure 2-2. This design may have its origins in an engineering contractor or an operator who had to improvise to get a level instrument installed in an operating plant. Either way, it must be recognized that the design will not provide accurate level readings.

Figure 2-2 Example of improper level instrumentation.

c02f002

8. There is often failure to use a methodical approach. While this limitation is closely allied with the previous one, it points out a need to structure even the best application of engineering principles. This structuring step is necessary to allow one to define which of the engineering principles are most appropriate. The failure to use a methodical approach could lead one to hypothesize erroneously that a fractionating tower had a plugged tray and that that was the cause of a high-pressure drop. In fact, the problem might well be associated with a change in internal vapor and liquid loading, buildup of an impurity that boils between the light key and heavy key, foaming caused by a trace impurity, or improper assumptions regarding the tower’s loading point.

9. The whole picture is often not seen. The problem solver who fails to use a methodical approach is vulnerable to arriving at the wrong answer because he fails to see the whole problem. There are often verbal clues which can hint that the problem solver is failing to see the whole picture. Some of these clues are comments such as, That’s a mechanical problem, or, The laboratory is wrong again. While these statements may be valid, they are often indications that the problem solver is excluding essential pieces of data. It should be noted that someone using the methodical approach is less vulnerable, but still subject, to this limitation.

10. There is often an over-dependence on history. While a historical database is a mandatory prerequisite for successful problem solving, the database should be used to define deviations rather than a repository of answers. The statement, The last time that this happened, it was due to … must always be tested by data analysis.

As described earlier, Figure 2-1 shows a typical industrial problem. Several of the limitations discussed above are apparent. The problem was certainly complex in that it could be caused by conditions in either the polymerization or the extrusion processes. There appears to be both a lack of a methodical approach and an inadequate application of engineering principles. In addition, while only a limited amount of data is present in Figure 2-1, the problem solution appears to be only historically based. There is no evidence that a hypothesis was developed and tested with a plant test. Was the problem solver seeing the entire picture? For example, was the independent variable polymerization production rate or extrusion rate? Was the independent variable production rate or residence time (the inverse of production rate)? Perhaps the confusion of the problem solver is illustrated by the figure, which shows the voids on the x-axis normally reserved for the independent variable.

3

SUCCESSFUL PLANT PROBLEM SOLVING

3.1 INTRODUCTION

Before beginning a discussion on how one conducts successful engineering problem solving, perhaps a definition of the activity is appropriate. Engineering problem solving is defined as the application of engineering principles to allow discovery, definition, and solution of plant operating problems in an expedient and complete fashion. The discovery and definition phases of problem solving are often ignored or considered obvious or unimportant. However, these phases prevent small problems from growing into large problems and allow the problem-solving phases to be done in an expedient fashion. Finding the problem involves sorting through the mass of laboratory and process data to uncover deviations that may only be a slight departure from normal, but which have the potential to grow into large deviations. Defining the problem involves developing a quantitative description of the problem specifications.

Successful engineering problem solving will always involve the following:

A daily monitoring system.

A disciplined (not intuitive), learned (not inherited) engineering problem-solving approach.

The ability to distinguish between problems requiring technical problem solving and those only requiring an expedient answer. The ability to determine how detailed a technical analysis should be is also required to efficiently solve plant process problems. This is later referred to as optimum technical depth.

3.2 FINDING PROBLEMS WITH A DAILY MONITORING SYSTEM

In order to successfully find and define problems, the problem solver must obtain and maintain a historical database. The database can be maintained by using several different sources. The managerial objective will also be important. The managerial objective is defined as the goal that management has defined for the particular process. This goal will vary depending on the age of the process, staffing of the location and the value added by the process to name a few. Table 3-1 shows a grid of both managerial objectives and sources of data.

Table 3-1 Sources of historical data

c03t01625iy

a The concept of key variable retention involves retaining the graphs or delta data graphs of only the key variables, whereas volume retention involves a data source that relies on maintaining values of every variable.

b Delta data graphs are the difference between actual values and a theoretical or established value. An example of such a plot is shown in Fig. 3-1.

As an example for the use of this table, assume that a well established process is producing a commodity chemical. As a general rule, a low value is added to commodity chemicals. That is, the difference between the product revenue and the cost of production is very small. Management might elect to staff this operation so that the organization could only respond to established significant problems. Thus the managerial objectives might be characterized as Minimizing Routine Work and Maximizing Variable Retention. In this case, the number of process variables to be retained would be maximized. As shown in Table 3-1, Computer Data Storage would be the desired source of data to fit this objective. If a problem developed, the problem solver could then go back and use the stored data to attempt to resolve the problem. He might find this difficult due to the vast amount of data that must be analyzed. In addition, the data sources entitled Communication with Hourly Workers and Visual Observation of Field Equipment would likely not be available since people’s memory might have faded and changes might have occurred in the field equipment.

On the other hand, if the process being considered is an unproven process and/or is a high value added process, management might elect the objective of Maximize Finding of Hidden Problems. In this case, the problem solver would use Delta Data Plots, Communications with Hourly Workers and Visual Observations of Field Equipment as his data sources. It is likely that the main source of historical data would be the trend graphs or delta data graphs. Of course, in this case, the computer would still be used to store all process variable data. However, it would not be the primary source of data for the problem solver. While this objective allows for finding problems quickly, it likely will require more technical and/or operations staffing.

In the two cases cited above there are implicit assumptions. In the case where the managerial objective is Minimizing Routine Work and Maximizing Variable Retention, the implicit assumption is that essentially all process problems that occur can be readily solved without a detailed problem analysis. In the case where the managerial objective is to Maximize Finding of Hidden Problems, the implied assumption is that essentially all problems will require a detailed problem analysis.

If graphs are to be used in any of the cases shown in Table 3-1,