Friday, June 24, 2016

System and Human Dimensions of Industrial Engineering - H. Harold Bass

Definition of Industrial Engineering by Narayana Rao KVSS: Industrial Engineering is System Efficiency Engineering and Human Effort Engineering.

Excerpts From

H. Harold Bass
Supervisor Research and Development Group
Industrial Engineering Division
Eastman Kodak Company, Rochester, New York
1963, University of California

The publishers have given permission to all to include the articles as needed provided reference is given.

 It has seemed to me that industrial engineers tend to get smothered in their growing body of techniques and perhaps lose sight of the ends towhichthese techniques should be directed.

If there is anyone theme which has persisted throughout the history of industrial engineering it
is the theme of improving Organization Performance.

There are a number of dimensions along which breadth of industrial engineering practice could be
discussed. I should like to emphasize two dimensions which I shall call the ''Systems Dimension''
and the ''Human Dimension."

I. The Systems Dimension

It seems that there has been one characteristic of traditional industrial engineering practice. Our
approach for many years assumed that we should concentrate on unit operations. Despite admonitions
such as ''look at the preceding and succeeding operation" we have by and large tended to optimize Organizational Performance piece by piece.

 The Systems' concept teaches that the sum of optimal parts is not necessarily equal to an optimal whole. Interactions between units or sub-systems and environmental variables can affect the performance of a system. The effects of this are only obvious when we grasp the system as a total system.

Linear Programming has enabled us to assess the total system of products and machines,
take into account the many inter-relationships of production rates and costs and arrive at a machine
loading program that is within the specified restrictions and is optimal for the total system.
Problems involving a large number of products and machines rapidly tax the capacities of the
most modern computers; however, efficient algorithms are capable of solving problems with as many as 300 products and 18 machines.

The linear programming approach has proven useful not only as a control in assigning products
on a weekly basis, but for long range planning of machine requirements such as;
1. Determining where to place development effort to overcome technological restrictions
which limit the capability of products to be produced on certain machines.
2. Determining the economics of adding additional capacity which is generally
more efficient than present equipment.
3. Determining the best allocation of products for special situations such as extended
periods of machine downtime for maintenance or design changes. This is just one example of work done in the Systems Dimension. Actually, the technology in this area has developed rapidly and can have significant influence on our goal of improving .Organization Performance.

II. The Human Dimension
For the Human Dimension, I'd like to cover two areas; Work Physiology and Organizational
Behavior.  Let's take Work Physiology, first.

A. Work Physiology

In the year 1903, Frederick W. Taylor in his book Shop Management stated, "One of the
most difficult pieces of work which must be faced by the man who is to set the daily tasks is
to decide just how hard it is for him to make the task." (end of quote) This, then, at the
time of Taylor was a decision under conditions of uncertainty and not a measurement. And,
judging by the labor disputes which arise over this question, we seem to be no nearer the point
of being definitive about this question than was Taylor.

What is work effort? Is there an objective way of measuring this effort? Effort can be defined
as an exertion of power or energy and this can be measured and quantified. Energy cannot
be measured by measuring time, but it can be measured by measuring such physiological
phenomena as heart rate, oxygen comsumption and respiratory volume.

The industrial engineer has implied consideration of these physiological phenomena in his concern with fatigue allowances and rest pauses, but even these allowances have generally been based upon time criterion rather than physiological measurements.

Today the industrial engineer has available to him, thru his own and other disciplines, the knowledge
and capability necessary for making quantitative determinations of many of these factors involved
in studying man at work. In the future, to successfully fulfill our role of studying man at work and to integrate him into the organizational system, we need to know the real demands which are being placed upon people. In a sense, we need to know their tolerance to physical and psychological loading.

While as engineers, we wouldn't think of regularly exceeding the design capacity of a production
machine, neither should we exceed the design capacity of the human operator. Knowing this
design capacity has definite humanitarian and economic value.

The optimum work situation is when the work capacities of an individual are compatible
to the work demands of the job. If we underload the individual, the situation is obviously inefficient
and costly. Although it is less obvious, the reverse is also very costly. The resolution of
grievances over work rate, working conditions, and the costs of on-plant medical care, as well
as compensation, do not come cheaply.

While I have earlier singled out the area of effort determination, I don't wish to imply that
this is the only area in which knowledge of human capacities and capabilities is needed. Some other
work situations in which such knowledge is needed are; tasks involving maintenance of a performance level during monitoring and vigilance tasks, frequent decision-making associated with
rapid paced operations, and the integration of an aging industrial population into an increasingly
complex and rapid paced industrial enviornment.

To define and approach these problems requires an understanding and application of physiological
knowledge. engineers. If they are to be solved, we must seek the assistance of other professions
and disciplines. No one discipline is sufficient within itself to bring to bear all of the effort
that is needed. This, then, dictates the need for the team approach. Our work physiology studies at Kodak Park, which started about five years ago, developed from a common need of the Medical Department and the Industrial Engineering Division to better understand the physiological limitations and capabilities of people. The understanding of common problems that has developed between
these two divisions, in itself, almost justifies the effort which has been expended on these
studies. The real reward, of course, is in our growing ability to evaluate and quantify situations
which heretofore, from a job design standpoint, had to remain unknown.

Initially, our investigations were confined solely to the ''effort" or "energy expenditures"
aspect of work. This was natural for several reasons, namely:
1. High effort is more obvious to the observer of the industrial scene and,
therefore, demands more attention.
2. While it is fractionated and scattered, much work has been done and reported by other investigators relative to "'energy expenditure", which is a measure of effort.
3. Instrumentation has been developed such that it is now practical to measure this variable of "energy expenditure" on the industrial scene.
Energy expenditure is measured by the indirect calorimetry method; that is, respiration
and oxygen consumption are the variables which are measured and converted to energy. Combined
with heart rate, these represent the physiological responses which we feel are necessary for accurately assessing high effort industrial jobs.

Physiological measurements in conjunction with time study now provide us with an insight
into industrial job design problems which is not obtainable in any other way. Using criterion
relative to energy expenditure, we can now assess jobs prior to the installation of a new work
standard or job design. We are in a better position to determine in terms of time and energy
what the job requires, the frequency of rest breaks, the necessity of providing auxiliary labor
saving equipment or the need for re-engineering the job completely. In situations where
the manufacturing process is not amenable to change, then the same physiological measurements
help us to select persons with the physical capacity demanded by the process. Most preemployment
medical examinations do not completely provide this information.

This new approach to designing industrial jobs has been successfully employed in many
types of jobs ranging from the handling of containers in a darkroom cold storage area to the
loading of box cars. A most rewarding use of it involved the pre-evaluation of a proposed piece
of production equipment. The work physiology studies indicated that additional materials handling
equipment was necessary if we were to obtain the anticipated increased production. The
nature of this materials handling equipment was such that installation at a later time would have
caused an extensive shutdown with the resultant loss of production.

With the measurement of and utilization of energy expenditure as a factor in job design, we
feel we are just beginning our work physiology studies. Energy expenditure is just one facet of
the problem. Other physiological phenomena of people may be studied so that we can integrate
them into job systems which take advantage of their capabilities and do not aggravate their
limitations. The result will be mutually beneficial to the individual and the company. The
second part of our Human Dimension is Organizational Behavior.

B. Organizational Behavior

I am using this term to refer to the behavior of people in an organizational or industrial setting.
As an area of knowledge, among other things, it refers to the reasons why people work or don't work, decide or don't decide to perform so as to achieve the objectives of their organizations.

If the concept of "Organizational Behavior" seems remote from industrial engineering to you, let me say that an incentive system, or any control system for that matter, is primarily designed to direct and influence the behavior of people towards organizational goals. As industrial engineers we are, it seems, in the business of designing systems to influence, direct and control human behavior, but we've never quite faced up to it in these very words.

The famous Western Electric Hawthorne studies of thirty years ago marked the beginning
of organized research into Organization Behavior. Since that time, studies in industry plus general
behavioral research have yielded information which promises utility to industry.

I think I can summarize the results of this research (and its utility to industrial engineers) this way.

You industrial engineers profess, in effect, a theory of management - a theory of how to organize men, machines, and materials so as to get the best results. The part of this theory which deals with men assumes that the performance of people will be best under situations where they are told exactly
what to do and how to do it, and are rewarded with money in proportion to performance.
Your way of doing business rests on certain behavioral assumptions.

''To put it another way, you are hipdeep in designing systems for influencing behavior and you make almost no use of the collective scientific information about the behavior of man. The assumptions which support your practice are not all wrong; they are just not complete nor up to
date. You need, first, to realize that you are deeply involved in influencing people's work attitudes, second, that you do this based upon certain assumptions, and third, that there is a good deal of information available which would alter and improve these assumptions."

You might assume that under the proper conditions, they will actually find personal satisfaction in
working towards your objectives.


In the short time left, I can only outline the manner in which we at Kodak Park are trying to
answer this challenge. In the first place, we have acquainted ourselves with the research
which bears on the problem. We have tried to integrate this to the best of our ability and reduce
it to the probable effect it may have on our practice. The following specifics are indicated:

1. Job Design

Instead of simply designing operations from the point of view of the optimum technical system, we think there are gains to be made in considering the nature of the jobs which people will do.
The usual industrial engineering criteria for job design stress extremes of task specialization. The consequences tend to be meaningless jobs. That is, jobs in which the individual has difficulty
seeing the relationship of his function to a larger whole. A version of what has been called Job Enlargement is called for. This is not just a matter of adding functions to a job, but adding a set of
functions which will comprise a set of activities leading to accomplishment of a visible objective. The activities making up a job should be examined to see if there has been a tendency to
remove the thinking functions and specialize them in other persons. Taken as a whole, we should endeavor to design jobs such that people have a maximum of control of the variables which
lead to end objectives.

2. Goal Orientation - Information Systems

A natural consequence of the over division of labor has been to focus the attention of individuals on very minute goals such as pieces per hour. We believe that industrial engineers should re-examine their approach to the goal setting function which is, after all, what time study has led to all these
years. People, it seems, do not behave on the job as isolated individuals. Many jobs are parts of a system and depend for success upon a high degree of interdependency of people. We are examining the structuring of goals to see what beneficial effect there is in providing the individual a perception
of his contribution to system goals. This takes the form of specifying job goals in terms of end-results and also in terms of the contribution of job level goals to system goals. Individuals are kept informed of system objectives, current progress of the system and any contemplated changes in objectives.
In effect, they are kept "in the know" about objectives and progress of the unit as well as their own job goals. In effect, we are trying to enlarge the focus of the individual relative to end objectives. By giving his more control through Job Design and overall goal orientation, we think his performance
and personal satisfaction will both increase.

III. Compensation or Incentive Systems
For many years, industrial engineering activity has been closely identified with incentives. The classical incentive approach stresses the closest possible relationship between pay and rate-of-output performance. As any of you who have administered an incentive system know, you have to take the bitter with the better. There are a number of practical problems or dysfunctions associated with incentives. I shall not stress these, but will try to describe the more fundamental problems. If we are to believe the results of behavioral research, people work for the satisfaction of a number of human needs. Only some of these can be satisfied by money. The most serious indictment of classical
incentives is that they have pre-occupied us with money to such an extent that we have largely
overlooked other considerations. Such things as achievement, responsibility, recognition and
work, itself, are satisfactions and sources of motivation in themselves. Our problem, here,
is to retain some monetary incentive, some pay/performance relationship, but not to do it
in such a manner that it is seen as the be-all and end-all of motivation. We believe that
closer attention to Job Design and Goal Orientation, previously mentioned, is one way of
providing a basis for satisfaction in the job. There is nothing in motivational research to indicate that relating rewards such as pay to performance is unsound. How this is done seems to be most important, however. We think that money should be looked upon as an after-the-fact reinforcement, not the primary initial motivator of good performance. In contrast to classical wage incentives, which stress close, short term, hour by hour correlation of pay and performance, the shift from "motivator"
to "reinforcement" may be brought about by extending the time over which pay and performance
are related. In addition, by utilizing longer time periods, performance considerations like quality, versatility and dependability can be considered in terms of pay.

We may close by reviewing the official definition of Industrial Engineering as it appears
on the AIIE Journal: "Industrial Engineering is concerned with the design, improvement, and installation of integrated systems of men, materials and equipment; drawing upon specialized knowledge and skill in the mathematical, physical, and social sciences together with the
principles and methods of engineering analysis and design, to specify, predict, and evaluate the results to be obtained from such systems."

If we are to believe this definition as a statement of what industrial engineers do, then we must assume that in our practice we do, indeed, draw upon knowledge from the social sciences. (But industrial engineering has not effectively drawn research conclusions or principles from the social or life sciences) . If we are to be designers of integrated systems of "men, materials and equipment," and if the design activity is to be based upon specialized scientific knowledge, then we had better equip ourselves to do so, (this article specifically focuses on)  the social or life sciences.

No comments:

Post a Comment