Stronger, higher, faster

By David Mendonça

Course teaches human factors engineering through competitive sports.

A recurring challenge in the education of OR/MS professionals is the need for practice in deploying key tool sets across a variety of domains. This need is particularly acute in the area of human factors engineering (HFE), a field traditionally concerned with the design and evaluation of combined human/technological systems. However, foundational HFE research has historically been undertaken in domains (such as emergency response) that rarely intersect with those found in other OR/MS areas (such as finance). These strong ties to HFE research domains therefore imply high startup costs for students wishing to engage HFE’s core intellectual content.

One approach to addressing this challenge is to identify domains rich in HFE-related phenomena but also relevant to student interests. To this end, the Department of Industrial and Systems Engineering at Rensselaer Polytechnic Institute (RPI) recently launched a course in “Human Performance Modeling and Support,” a distinguishing feature of which is its focus on a domain of broad interest, good data and a burgeoning wave of scholarship: competitive sports.

This article discusses course structure and content and how it leverages student knowledge to achieve a somewhat paradoxical result: an understanding of HFE methods that is, to a large extent, independent of application domain.

Scope and Objectives

The course draws on the historical roots of HFE in OR/MS, as well as more recent research (see Interfaces, Vol. 42, No. 4). A particular focus is on methods for identifying and formalizing links between cognitive, physical and contextual determinants of human performance. The choice of the competitive sports domain is informed by three classic challenges for HFE education.
First, many HFE examples require highly specialized domain knowledge to decode, thereby draining time and effort from work on methods. Here, the breadth of sporting pursuits is balanced by shared concepts (such as rules of the game) with which students are very familiar.

Second, students should be able to deploy their tools effectively in a variety of domains, but opportunities for doing so in the classroom are limited. Here, the unique characteristics of each sport require students to develop effective techniques for moving between sub-domains (e.g., curling to bicycling). The wealth of good sports data supports their efforts.

A third challenge is related to the markets (and marketability) of the profession. The field of sports analytics is growing rapidly, translating into additional job opportunities for graduates. The sports domain also provides built-in measures for assessing the potential impact of their skills through better understanding of what makes a winning team.

Competitive sports provide a compelling domain to investigate and test core methods from HRE.

Competitive sports provide a compelling domain to investigate and test core methods from HFE.

Balanced against these positives are some possible pitfalls. First is the prospect that student attention will shift from methods to sports per se. Methods are therefore always introduced “sports-free” to maintain grounding in core scientific concepts – a perspective emphasized through the textbook, “Working Minds.” Second is whether students can apply these techniques to other domains. Selected examples from non-sports domains are therefore reviewed. Another concern is that less sports-oriented students may be at a disadvantage. They are told the opposite is more likely, as they are not burdened by potentially faulty assumptions.

The course content is grouped into three modules:

  1. Knowledge elicitation is concerned with the early stages of modeling: that is, with identifying a problem, the key factors that define it and the relationships among these factors. This module introduces interview-based techniques (such as concept mapping and the critical decision method) and methods for the analysis of these results (such as digital encoding of decision rules).
  2. Observation and measurement activities range from research on human-based methods to computer-based methods employing sensor-based technologies. The emphasis is on capturing measures associated with factors identified in module No. 1.
  3. Technology and training bridges measurement and performance support, including the use of technology for extending human cognition in training and performance situations. Feedback modalities such as data/information visualization are also covered.

In each module, extensive use is made of video and audio from a wide variety of sporting contexts, utilizing an expanding library of resources from YouTube and other sources. Students complete individual work on two projects to cement their knowledge, examples of which are summarized in Table 1.

education table 1

Table 1: Example projects

Course development involves considerable outreach to athletes and athletics organizations. Within RPI, these include RPI Athletics, which provides data and consultation for student projects. This year, students are working with play-by-play data from RPI hockey and football, combined with observational studies on the field of play. Outside RPI, the course draws external speakers including coaches, sports psychologists and technology developers. In a new initiative, this year students will visit the U.S. Olympic Training Facility at Lake Placid, N.Y.

Competitive sports provides a compelling domain in which to investigate and test core methods from human factors engineering. Students’ familiarity with the domain enables them to ramp up almost immediately on these methods, and to engage and critique associated data and models. They complete the course having followed the arc of modeling from problem identification and formulation, through to data collection, interpretation and presentation of results.

David Mendonça ( is an associate professor in the Department of Industrial and Systems Engineering at Rensselaer Polytechnic Institute.