The Engineering of Human Work

“Human work engineering” is neither a major in any university nor a job title I have ever encountered. As a specialty, it would integrate content currently filed under Human Factors, Ergonomics, Safety, Human-Machine Interfaces, Usability Engineering, Mistake-Proofing, and Jidoka into a consistent approach to production and service delivery.

But wait! Isn’t it what Industrial Engineering (IE) was supposed to be?

Whatever became of Industrial Engineering?

The pioneers of IE were focused on human work.  Taylor, the Gilbreths, or Gastev focused on individual operations. 50 years later, IEs like Taiichi Ohno and Shigeo Shingo worked on processes with multiple operations. It is, however, not what IE has come to mean today.

In the 2019 Industrial Engineering Body of Knowledge, the IISE outlines the discipline, as it sees it. Of its 12 chapters, only 3 are about the engineering of human work:

  1. Work Design and Measurement
  2. Ergonomics and Human Factors
  3. Safety

The other chapters describe technical approaches that are not about human work. They occasionally discuss people but only about labor requirements, support for change management, or the need for collaboration.

The Evolution of IE

Why IE has drifted away from its original focus is an open question, because the design and execution of human work are as important today as 100 years ago. Technology has changed and the challenges are different but there are still challenges. 100 years ago, the work of driving a train was busy, physical, and dangerous. In today’s high-speed trains, the challenge is to keep the driver alert and able to respond to emergencies, much like an airline pilot.

The original IE had limited success. A lasting legacy of Taylor is the image of  IEs on the shop floor in a lab coat wielding a stopwatch and a clipboard with the goal of speeding up the line. The workers resisted and, over the decades, were successful in blocking these efforts.

Frank and Lillian Gilbreth recorded operations on film to analyze and make them easier to execute. They were ahead of their time, as the costs in the 1910s were too high to do it systematically. Today’s technology has changed the game. With smartphones, everyone can now make video recordings of operations and review them immediately.

Peer pressure

Within engineering schools, mechanical, electrical, and chemical engineering students looked down on IE students, as short of “real” engineers. Never mind that the business world disagreed, offering IE graduates higher salaries and faster promotions than other engineers. Perhaps, companies valued the emphasis on people and, rather than 2nd-rate engineers, saw IEs as potential managers who could add, like Jack Swigert on Appolo 13:

Still, university IE departments had to make themselves attractive to students choosing majors. For this purpose, image mattered. You could boost respect for IE by packing the curriculum with math as esoteric as that of the “real” engineers. Operations Research (OR) filled this purpose, and delivered complexity on a par with fluid dynamics, gene sequencing, or neural networks. OR is about operations but the closest it gets to human work is task assignment. It doesn’t go into what Crispin Vincenti-Brown summarized as “what happens when the guy picks up the wrench.”

Name Changes

The IE departments changed their names from just “Industrial Engineering” to “Industrial and Systems Engineering” (ISE or ISyE), or “Industrial Engineering and Operations Research” (IEOR). It was not just rebranding. It reflected actual changes in curricula, in which the original IE subjects took a back seat.

At the turn of the century, Stanford University merged IE with OR and Engineering-Economic Systems into the current Management Science & Engineering (MS&E) department. “Management Science” is synonymous with OR and, as an applied field, is truly Engineering rather than Science. In other words, Stanford’s reorganization completely erased IE from the department’s name; it’s all OR. Out of 30 faculty members at MS&E, only four are making a last stand in a center called Work, Technology, and Organization.

Access to Human Work and Simulations

You can only observe human work in workplaces. In mechanics, electronics, chemistry, or even robotics, students can experiment in a school lab; for human work, they can’t. A university department teaching the engineering of human work needs industry partners to open their doors to students and host projects. This subject is actually a good fit for an apprenticeship system, where students alternate between extended internships and classroom training.

The closest you can get to human work within a school is through simulations, and suppliers of simulation software are happy to provide their products at heavily discounted prices for students to learn and get hooked on. While simulations are useful tools in business operations, schools overemphasize them because they are more easily accessible than actual workplaces.

As a result, students will confuse simulations with real things. Simulations are always simplifications, and most valuable when their user has access to the actual system. In addition, the development of simulations is no small effort — even with advanced tools — and, in a real workplace, it is not a foregone conclusion that it is worth doing.

For example, in the following video, R3DT’s VR-based assembly simulator lets you experience the same workstation from the perspective of operators who are 5-ft and 7-ft tall. To grab a virtual object, you must first touch it with your index finger and your hand closed. Then a halo appears around the object and you can grab it but, unlike real hammers and pliers, the virtual objects have no weight.

The Operations Research Takeover

About 20 years ago, Hormoz Mogarei and I presented our work to the Ph.D. students in the IE department at Stanford University, in the hope of generating interest. Their unanimous reaction was that they hadn’t gotten this far in school to get involved with low-level work like operator job design. Their focus was OR. Today, as reported above, the IE department at Stanford no longer exists.

When I spoke at the APIEMS conference in Kitakyushu, Japan, in 2009, I was the token “Lean expert” for that year. Other participants told me there was always one, and it was my turn. Most of the talks were by academic researchers, about using genetic algorithms to form manufacturing cells, schedule production or otherwise optimize some metric. In factories, however, I had never heard of genetic algorithms before and I haven’t since.

Prof. Mitsuo Gen, author of several books on the subject, told me then that young researchers could get funding for genetic algorithms at the time. 5 to 10 years earlier, clustering had been hot but no longer was. Today, genetic algorithms appear along with simulated annealing, tabu search, neural networks and other techniques as part of a class called metaheuristics. Genetic algorithms are also described as artificial intelligence.

The IE Identity Crisis

Industrial Engineers most often cite Maynard’s and Salvendy’s handbooks, both last updated in 2001. The most recent handbook is Badiru’s, whose 2nd edition came out in 2013. Maynard’s represents the perspective of American consultants, and it devotes 184 pages to defining what the profession is and explaining its past, current, and future roles.

Salvendy’s was written primarily by academics and has 60 pages on these topics. In Badiru’s 2013 handbook, also written by academics, it is down to 22 pages. By contrast, Dubbel’s Handbook of Mechanical Engineering has none. It goes straight into the statics of rigid bodies without a word about what a mechanical engineer is. The authors seemingly assume that no reader would ever ask.

The need to devote so much ink to defining the profession reeks of an identity crisis.

Current Definition of an IE

In his introduction, Badiru quotes the definition given by the Institute of Industrial & Systems Engineers (IISE):

“Industrial engineering is concerned with the design, improvement and installation of integrated systems of people, materials, information, equipment and energy. It draws 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.”

But he then admits that “some practitioners find the definition to be too convoluted.” Besides being convoluted, it is also vague and sets unclear boundaries. That I am concerned about climate change, for example, doesn’t mean I know how to prevent it. An engineering discipline is about more than a concern.

As for the boundaries, there are few artifacts that cannot be described as integrated systems of people, materials, information, equipment, and energy. A lump of coal is clearly not a system but a nail, arguably, is one, integrating a head, a shank, and a point into a system that people use with a hammer. While this definition is purportedly about industrial engineering, it says nothing that is specific to it. Everything it says applies instead to systems engineering, as described by INCOSE.

It fails to answer the question of what IEs are and how they differ from other engineers. There used to be a clear and concise answer: they were the engineers of human work.

The Scope of Human Work

Besides paid activities, “Human Work” also includes household chores or volunteer activities. What it does exclude is entertainment. People work for pay on production lines, in hospital operating rooms,  in offices, or at supermarket check stands;  they cook at home, take out the trash, and answer the phones at fundraisers, for no pay. All of it is still work.

Playing video games or exercising, on the other hand, is not work, but these activities still deserve attention because their technology occasionally has crossover value at work. Gaming interfaces can be repurposed in training, simulation, or supervisory control; video analysis tools for athletes have a dual use in production operations.

What is the Engineering of Human Work

The engineering of human work covers everything that is physically needed for humans to do the work. At each workstation, it includes many components:

  • Machines
  • Tooling
  • Materials handling and presentation
  • Check gauges
  • Conveyance systems to bring the work in and take it out
  • Process instructions
  • Production schedules

At the facility level, it includes the following:

  • Pathways from the street to the production line at shift start and end
  • Lockers for personal items
  • Cafeterias and break rooms
  • Conference rooms
  • Restrooms
  • First aid

Done well, this kind of engineering makes it easy for people to do what they do often, keeps them safe, does not make them wait, and enables them to improve work methods, also known as Kaizen.

Management and Technology

Undoubtedly many readers will object that engineering tools are not what matters and that the work performance is determined by management,  in particular by leadership and the culture it fosters. Provide strong leadership, they will say, and technical matters will take care of themselves. My experience, however, tells me that management, leadership, and culture are no substitutes for engineering know-how. You don’t achieve much if you try to improve business operations without addressing the engineering dimension of its members’ work. Conversely, you don’t achieve much by engineering alone. You need to cover all the bases.

Professionals, as humans, tend to undervalue specialties other than their own. When you envision a factory or a business unit as a system, you see multiple, equally important dimensions. Even if your background is in social sciences, business administration, or the priesthood, you need to understand that engineering, particularly of human work, cannot be given short shrift.

#humanwork, #engineering, #management, #industrialengineering, #operationsresearch, #IE, #OR, #usabilityengineering, #mistakeproofing, #pokayoke, #ergonomics