The First 50 Years of the Department of Industrial and Operations Engineering at the University of Michigan: 1955–2005
Skip other details (including permanent urls, DOI, citation information)
: This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License. Please contact [email protected] to use this work in a way not covered by the license.
For more information, read Michigan Publishing's access and usage policy.
6. Building Centers and Programs
Six important collaborative efforts were developed through the leadership of industrial and operations engineering (IOE) faculty members over the past 50 years:
- Information Systems Development and Optimization Systems (ISDOS)
- The Center for Ergonomics
- The Dynamic Systems Optimization Laboratory
- The Engineering Global Leadership Program
- The Tauber Manufacturing Institute
- The Financial Engineering Master’s Program
- The Hospital Systems Engineering Program
Each of these programs had unique qualities and had a significant impact on the education and research within the department. In many ways the department’s continued high national ranking can be attributed to the leadership necessary to develop and operate these collaborative programs. In the following sections each of these programs is described.
6.1. Establishing a Major Industrial Engineering–Oriented Information Systems Organization—ISDOS
6.1.1. Background of ISDOS
As described earlier, in the ’60s the department relied on Richard Wilson to develop and instruct students on how to use digital computers to create analytical operations research models. In addition, Dean Wilson drew upon his expertise in the iconic American Airline’s Sabre reservation system to teach the use of various types of programmable digital computers in manufacturing and signal processing, as well as in implementing the Monte Carlo system-simulation method. Bert Herzog, who had joined the department in 1965, was directing research and teaching in computer graphics and networks. In 1966, these three were joined by Ed Sibley from MIT, whose expertise was in management information systems. Dan Teichroew had begun his work on Problem Statement Language and its supporting software, Problem Statement Analyzer at Case Western Reserve in the early ’60s. In 1968, Teichroew joined the department as its chair. Soon after, in 1970, Alan Merten was recruited from the University of Wisconsin’s Computer Science Department.
Thus, by 1970, six faculty members were instructing in and researching the use of digital computers and computer language development to model and analyze complex information technology systems.
6.1.2. Overall Focus of Research in ISDOS
Teichroew’s concern in the late ’60s was that information systems lacked the equivalent of a manufacturing blueprint—a coherent, complete plan of what is going to be built that ensures that the construction will meet all the requirements. So the initial thrust of the ISDOS project was to develop a mechanism for describing an information system and its requirements. The mechanism was linguistic because, in the 1960s, computer graphics were just starting to become available. In addition, personal computers had not yet been developed as local workstations, so all development was on a mainframe computer.
The blueprint modeling mechanism that the ISDOS team developed, initially at Case Western Reserve University in 1967, was called the Problem Statement Language (PSL). PSL modeling was based on an extended entity-relationship model. PSL statements describing an information system and its requirements were presented to a repository managed by the Problem Statement Analyzer (PSA) software. PSL statements were synthesized as they were presented and only checked for syntactical correctness and lack of contradiction at the time of entry, but not for completeness.
The PSL statements could be added into the model by different analysts at various times, so a system could be described in modules or pieces (similar to the concept of modular, agile programming). Then, when the PSA analyzed all the pieces, the model could be refined and completed by discovering gaps, resolving design issues, identifying overlaps, and so on. PSA could generate reports on different aspects of a system model from the contents of the repository and check for model completeness. When the system description was deemed complete, it could be invoked to produce a composite description and specifications for the information system under development.
6.1.3. Innovative Strategy for Funding ISDOS Research
Teichroew proposed that ISDOS should be funded by contributions from potential users, he termed them “sponsors.” who paid a modest yearly subscription fee. In return, the sponsors could attend yearly workshops at Michigan and selected off-site locations. Sponsors were given the research products to take back with them and apply to their work. (This necessitated the implementation of PSA software across various environments, such as IBM, Honeywell, Burroughs, CDC, and DEC.) Sponsors were asked for feedback on the utility of the ISDOS development products. This resulted in unheralded techniques of storing the software in a database and mapping it into various environments. Some sponsors wanted PSA reports to match the framework of their documentation standards, and the resulting ISDOS product was the Automated Documentation System, which formatted PSA reports to an organization’s standardized content.
ISDOS products were developed by IOE graduate students (five to 10 at a time) guided by faculty and senior researchers. Results were published in books and national journals. For example, the key article “The Automation of Systems Building” was published in Datamation in 1970 and later included in the books Systems Analysis Techniques (edited by J.D. Cougar and R. Knapp, John Wiley & Sons, 1974), and in Classics in Software Engineering (edited by Edward Nash and Ed Yourdon, Prentice Hall, 1979). Yourdon and Teichroew developed a relationship that resulted in PSL being used to formalize Yourdon’s hitherto informal Data-Flow Diagrams. Worldwide exposure from these publications stimulated interest in ISDOS work from foreign organizations. Teichroew was invited to make presentations about his work and consequently drew foreign sponsorships for ISDOS.
The sponsorship model may look very familiar to today’s researchers, but it was innovative and unusual at the time. It had the beneficial side-effect of mitigating conflict-of-interest issues between researchers and the people who funded them and reducing dependency on one or two major sponsors. Sponsors who did not have or did not want to spend a substantial amount could still benefit from the innovative research and development products. Essentially, Teichroew had established and was operating one of the first consortiums in the University, and at its peak in 1982 ISDOS was providing more than $1.1 million in sponsored research funds to the IOE Department (the equivalent of about $2.6 million in 2014).
6.1.4. Major Users and Supporters of ISDOS
United States sponsors included the US government, including many Department of Defense organizations and contractors as well as commercial organizations. The Social Security Administration was a major user of PSL/PSA and required all of its contractors to use PSL to describe the information systems they were proposing to develop.
Military users included the US Navy. Groups such as the Naval Ocean Systems Command, Naval Underwater Systems Command, Naval Ship Research and Development Center, Naval Surface Warfare Center, and Naval Weapons Center at China Lake used PSL/PSA. When the Navy wanted to procure a computer-aided laboratory, it developed the request for proposal, that is, its requirements, using PSL/PSA and the Automated Documentation System. Various groups in the US Air Force also sponsored the research. A major request for proposal released by the Air Force was in the form of PSL statements. A special version of PSL/PSA was developed for the Electronic Systems Division at Hanscom Air Force Base in Massachusetts called URL/URA under the MULTICS operating system. Corresponding courses were also designed.
In the corporate arena, Raytheon, Rockwell International, Computer Sciences Corporation (CSC), and TRW were among contractors using PSL/PSA; TRW developed a specialized version of PSL/PSA that they labeled RSL/REV under an umbrella called SREM. IBM obtained a corporate-wide license for PSL/PSA. Other commercial users included the Boeing Corporation, National Cash Register (NCR), the Upjohn Company, Mobil Oil, Diners Club, Pacific Gas and Electric in Southern California, Honeywell, Burroughs, CDS, DEC, and computer-aided software engineering tool vendors, such as Index Technology of Cambridge and the NASTEC Corporation of Southfield, Michigan. Various banks also used PSL/PSA. Among the foreign users were the August Thyssen Group in Germany, Dutch Post Telephone & Telegraph, British Rail, ARAMCO, and L. M. Ericsson.
6.1.5. ISDOS Becomes a Commercial Spin-off
To help showcase to the taxpayers of Michigan how university research contributed to the state’s economy, the ISDOS project was selected and spun off (around 1983) as ISDOS Inc., headquartered in Ann Arbor. Elliot Chikofsky was chief of marketing. ISDOS Inc. was eventually bought by Methods Works Inc. of Canada. It, in turn, was sold to a British company. Richard Welke, then principal of Methods Works went on to Georgia State University, where he was appointed professor of computer information systems and director of the Center for Process Innovation.
During the transition years to a commercial company (1983–1985), much of the research work was left in the IOE Department. To avoid confusion, in 1983 the department laboratory was renamed the Program for Research in Information Systems Engineering (PRISE). Some of the sponsors of ISDOS continued to support Teichroew’s efforts in PRISE, which included supporting several PhD students. Unfortunately, at the time the University General Counsel’s Office did not have a clear policy on how a sponsor could divide funds between the UM’s PRISE Laboratory and ISDOS Inc. without creating a conflict of interest for Teichroew. The result was that PRISE was not well funded after 1987. Also, as computer science departments emerged in various academic institutions, including UM, the research work on high-level program languages shifted away from industrial engineering (IE). By the early ’90s the PRISE Laboratory had closed.
6.1.6. Influence on ISDOS Graduates
ISDOS influenced many graduate students. Jay Nunamaker, a student of Teichroew’s who worked on an early version of PSA, went on to become dean of the College of Business and Public Administration at the University of Arizona. One of the 1973 IOE PhD graduates, Hasan Sayani (whose major adviser was Ed Sibley) formed an informal “ISDOS East” at the University of Maryland. With help of a cognitive psychologist and curriculum designer, Cyril Svoboda, PSL/PSA courses were revamped. Special courses meant for the larger population of PSA report “readers” were designed, along with corresponding, more detailed courses, for the “writers.”
Sayani was also instrumental in having PSA implemented on the UNIVAC computer system at the University of Maryland; this was then implemented on the Social Security Administration’s computer system. Sayani implemented PSA on the HP/UX systems and sold it as a turnkey system (e.g., to the Naval Surface Warfare Center). Several associated systems were developed by the ISDOS East group as well; among them, a reverse engineering system (REVENGG), a logical database design system, and the Automated Documentation System, which incorporated standard text with PSA reports. The Automated Documentation System dynamically inserted PSA reports within free-form text. Two-way bridges were developed between PSA and computer-aided software engineering tool products.
Other IOE PhD graduates who were supported and guided by Dan Teichroew, Ed Sibley, Alan Merten, and Tony Woo during the ’70s and ’80s were: William Ash (’71); Milton Drott (’73); David Carlson (’75); Koichi Yamaguchi (’75); S. Navathe (’76), who was named an Association for Computing Machinery fellow in 2014; David Johnson (’81); Kyo Kang (’82); Timothy Thomasma (’83); Sung Yong Shin (’86); and Hyun Lee (’88). All went on to have prominent positions in the information technology world.
6.1.7. ISDOS Work Overtaken
Although many sponsors and clients found PSL/PSA an exciting advance during the ’70s and ’80s, attempts at commercialization were ultimately stymied. Though PSL/PSA principles were integrated into computer-aided software engineering tools, capabilities were overpromised, eventually discouraging sales. A common organizational issue also arose with management impatience with the rigor and analyst time required to produce good information technology requirements and specifications. And though Dan Teichroew continued his work within the IOE Department until his retirement in 2001, Hasan Sayani invited Ed Sibley to join the Information Systems Management Department at Maryland in 1973. When that was disbanded at College Park, Sibley left for the Department of Information and Software Systems Engineering at George Mason University, where he is now a professor and eminent scholar. Sayani went on to direct the graduate program in software engineering at University of Maryland University College, where he drew upon his prior experience to develop a commercial Semantic Database Management System—CaMERA. In 1975, Bert Herzog left UM to join the engineering and computer graphics faculty at the University of Colorado. He later returned to UM in 1987 as director of the Computing Center. Alan Merten transferred his appointment to the UM School of Business Administration in 1976, and after appointments at Cornell University and University of Florida, he went on to become president of George Mason University in Virginia (he recently retired). Tony Woo, who joined the IOE Department in 1977 and was instrumental in the development of models and computer algorithms to guide the geometric assembly of complex shapes, left in 1990 to join the IE department at the University of Washington.
6.2. The Center for Ergonomics
In 1979, the ergonomics faculty within the IOE Department requested that a designated center for research on ergonomics issues be established within the College of Engineering. After review by the College administration and the vice president for research’s advisory board, in November 1979 the regents of the University granted the request, and the Center for Ergonomics was created. Because of its scope of activities and its importance for more than 35 years, chapter 7 is devoted to describing the history of the center and its impact on the field of ergonomics.
6.3. The Dynamic Systems Optimization Laboratory
The Dynamic Systems Optimization Laboratory was formed in 1985 and is housed today in a computer lab on the main floor of the IOE Building. It was created by Professors Robert Smith, James Bean, and Jack Lohmann to serve as an applied research consortium that could unite the shared interests and activities of those faculty and students engaged in the modeling and analysis of problems involving dynamic, sequential decision making over time. The lab was novel in its establishment of a common space with shared computational facilities for operations research faculty and their students to engage with real-world industrial firms and governmental entities to tackle some of the most demanding problems confronting the world. Its primary emphasis was on improving the design and operations of complex systems. The lab’s original computational facilities were provided by a special grant from Bell Laboratories that specifically supported the lab’s work on establishing the existence and determination of planning horizons for long-term strategic planning. The National Science Foundation and General Motors provided many subsequent grants for this field of infinite horizon optimization, totaling roughly $1 million over the years. The PhD students at the lab whose dissertations were written in the arena of infinite horizon optimization and asset planning included David VanderVeen, who later became director of product development analytics at GM; Matt Bailey, who acquired the title of Howard J. Scott Professor of Management at Bucknell University; and Sarah Ryan, who became a professor at Iowa State University and an Institute of Industrial Engineers fellow. Also the National Science Foundation CAREER Awardee Allise Wachs, who later became president of Integral Concepts, and Tim Lortz, who is on the technical staff at Booz Allen, were early contributors. Two other notable PhD students whose dissertation research in infinite horizon optimization formed some of the theoretical foundations for the laboratory were Wally Hopp, who became the Herrick Professor of Business and senior associate dean in the business school at the University of Michigan, and Julie Higle, who later became chair of industrial and systems engineering at the University of Southern California.
One of the other recurring themes of the lab was the challenge of designing and managing the flow of vehicular traffic to relieve delays caused by recurring and incident-caused traffic congestion. The lab tackled this problem in the private and governmental sectors through dynamic route guidance and coordinated traffic-signal control. Stephane Lafortune from the EECS Department played a prominent role on the problem of coordinated signal control. This research was supported by the FHWA Intelligent Transportation Systems Research Center of Excellence at the UM. Participants included PhD students Karl Wunderlich, who later became a tech fellow at Globis Corp, and David Kaufman, formerly at AT&T Labs. The approach was to animate the vehicles and traffic signals within a very large game played by tens of thousands of players within a computer simulation that discovered the best cooperative strategies though a fictitious-play algorithm.
This fictitious-play algorithm was later generalized with the support of several National Science Foundation grants to orchestrate the design and operation of general complex systems. One application of fictitious play was the design of a dedicated production line, supported by grants from General Motors through the GM Collaborative Research Lab at UM. This application integrated revenue management and inventory control through their interplay within fictitious play. As PhD students, Archis Ghate, associate professor at the University of Washington, and Shih-Fen Cheng, associate professor at Singapore Management University, played central roles in this research. Another application, supported by a Multidisciplinary University Research Initiative (MURI) grant from ONR, provided an algorithm for minimum-time path finding for naval vessels in an evolving ocean wave field. Another MURI hrant, this time from the Army Research Office, for the design of low-energy mobile communication electronics for situation awareness, led to implementation of the hit-and-run algorithm developed by Smith within a large-scale simulation optimization algorithm for overall design of the mobile electronics system. Hit-and-run was later incorporated by Zelda Zabinsky and one of her PhD students working with Boeing Corporation, which resulted in a software suite for the design of composite airplane components. This software was later used in the design of the Boeing Dreamliner. Hit-and-run is often regarded as the most efficient platform for Monte Carlo generation within complex stochastic simulation samplers.
The lab’s funding for these activities was large scale for the field of operations research, numbering in the hundreds of thousands of dollars per year. Other PhD students who served as research assistants within the lab and worked in several of the research areas were Jeff Alden and Dan Reaume, who later became tech fellows at the General Motors Research Labs; Edwin Romeijn, who joined the IOE faculty for several years and later assumed the chairmanship of the School of Industrial Engineering at Georgia Tech; Alfredo Garcia, who became professor of industrial and systems engineering at the University of Virginia; and Irina Dolinskaya, who later held the William A. Patterson Junior Chair in Transportation at Northwestern University. Jeff Alden later won INFORMS’ most prestigious prize, the 2005 Franz Edelman Award for Achievement in Operations Research and the Management Sciences.
6.4. The Engineering Global Leadership Program
In 1992, IOE professor James Bean established the Engineering Global Leadership (EGL) program, a five-year honors program designed to strengthen the IOE educational program by addressing three recognized issues: (1) the need for young engineers (particularly new degree recipients) to develop and improve their communication skills to breach the engineering-business boundary, (2) the increasing requirement that graduates learn how to comfortably and effectively employ their skills and abilities when working within another culture, and (3) correct the lack of a formal honors program within IOE (and in the College of Engineering) that could recognize and promote individual excellence and global leadership skills.
The core curriculum of the EGL program addressed the first two of these gaps by combining the then traditional engineering curriculum with core courses in the School of Business Administration; the School of Literature, Science, and Arts; and selected foreign educational institutions. The former explicitly addressed leadership and organizational issues, and the latter exposed students to the languages, history, and customs of a student-selected non-US country or region of the world with existing or potential competitive and engineering presence.
The curriculum also required students to complete a synthesis team project that placed them within a specific and real industrial context, to which they had to apply their technical knowledge and develop their teamwork skills. This exposure of excellent students to organizations with operational and other engineering missions, and (perhaps more important) exposure of the organizations to these students, was an immediate success.
The EGL program was also an honors program, since the admissions (and subsequent academic performance) requirements were very high. Even though its added course requirements and constraints were time consuming, the benefits were clear, and from the very beginning the enthusiasm and accomplishments of the students were exemplary.
At first, the program led to two degrees after an average of five years at Michigan: a BSE and then an MSE in industrial and operations engineering. By the year 2000 the EGL program was made available to a small number of students in mechanical engineering and manufacturing engineering. (Over the following five years, under the guidance of then-EGL director Stephen Pollock, the program was further extended to students in all other College of Engineering degree programs, although the majority of EGL students still obtained their degrees in IOE.) In the mid-1990s, enrollment in the EGL program was five to 12 students per year, but by the early 2000s, after the incorporation of non-IOE students, EGL enrollments grew appreciably and then leveled off at about 50 per year.
The payoff of this innovative program was immediate: employers recognized that EGL students were not typical master’s degree graduates, and that they did not fit into the traditional engineering job role. In response, some companies actually created new positions and job rotation programs specifically for EGL graduate, to take advantage of their unique skills and talents. By 2005, similar programs were instituted in other universities, but UM’s IOE led the way with the first one that combined cross-disciplinary engineering-business education, requirements for understanding international cultural aspects of engineering, and honors recognition for academic and applied engineering excellence.
In the early days of the EGL program, student projects were sponsored by a wide variety of firms, including Allied Signal, Cummins Engine Company, DaimlerChrysler Corporation, Dell Computer Corporation, Ford Motor Corporation, Intel Corporation, and Lucent Technologies, as well as consulting firms such as A.T. Kearney, Bain & Company, Boston Consulting Group, Diamond Technology Partners, McKinsey & Company, and PricewaterhouseCoopers. By the turn of the century, student projects became integrated with the Tauber Manufacturing Institute (discussed in the next section), which provided a richer and more stable source of student projects sponsored by over 30 organizations each year.
6.5. The Tauber Manufacturing Institute
6.5.1. Early Years (1993–1999)
The concept for a multidisciplinary business and engineering enterprise at the University of Michigan was born in November 1991 as a result of a joint meeting between the College of Engineering and School of Business Advisory Boards. A Program Development Advisory Board—consisting of industry executives from 27 corporations—was assembled and held its first meeting in the spring of 1992. Through the course of many more meetings, the goals, core curriculum, and organizational structure of a new program named the Michigan Joint Manufacturing Initiative (MJMI) was launched in 1993.
IOE professor James C. Bean was appointed the codirector as part of his role in the College to broaden the scope of the undergraduate engineering curriculum beyond predominantly technical subjects. His leadership from 1993 to 1999 allowed the MJMI at the University of Michigan to be the equal of MIT’s Leaders for Manufacturing. Both programs were created at a time when Japan and other overseas rivals were challenging US manufacturing dominance in many areas, including the automotive industry. Peter Banks, dean of the College of Engineering, and B. Joseph White, dean of the Stephen M. Ross School of Business, were at the helm when the innovative joint institute took shape.
Business school professor Brian Talbot and IOE professor Chip White were asked to join Jim Bean in leading the program. They recruited Craig Marks, who had just retired from TRW. Marks served as the institute’s industry codirector from 1993 to 1999.
It is worth noting that the concept of such a joint educational program between two large and highly ranked schools was controversial at the beginning, but the success of the student team projects (described later) played a major role in the institute’s gaining acceptance. Students were already aware that a cross-disciplinary education was the future, and once they heard about the program’s details, they began gravitating toward the institute. US companies that were losing market share to overseas interests and other manufacturers were ready to invest in the creation of a new kind of employee—one with training in business and engineering.
A pivotal point in the existence of the MJMI came in 1993 when Peter Banks, who had strongly supported the institute, announced that he was resigning as dean of engineering. Some opponents of the institute were ready to let it dissolve at that point, but Joel Tauber, a UM alumnus and manufacturing executive from Detroit, gave the institute its second wind in the form of an endowment, and this led to a name change: MJMI became the Joel D. Tauber Manufacturing Institute (TMI).
TMI soon became one of the top multidisciplinary programs in the country. It was the only major manufacturing program that offered a full range of degree options for master’s and undergraduate students. In addition to normal degree requirements, TMI students completed an intensive leadership program in engineering and business that culminated in a team-based, 14-week paid internship with a high-profile manufacturing company.
Team projects were (and continue to be) at the heart of the institute’s philosophy, which focuses on teamwork, integration, and manufacturing success. Teams usually comprised two or three students from the School of Business and the College of Engineering, and the student team projects were designed to test students’ collaborative skills. Working as engineers and business managers, teams collaborated to analyze a current manufacturing problem and design and implement a solution that would improve the sponsoring company’s performance.
Though traditional manufacturing companies were key players in the early days of TMI, companies in electronics, pharmaceuticals, paint, and other industries came on board. In fact, Dell Corporation went on to hire a substantial number of Tauber graduates to work on its new supply chain strategies. Because Craig Marks was well connected with senior industrial leaders and knew that corporate America was looking for a new type of executive, he was able to provide TMI the business resources it needed for a number of years. He put together a team of industry people who were willing to work to help UM create a program that would shape the education of their potential new executives.
The open cooperation of industry executives created ownership of the new curriculum structure and resulted in additional financial support from the participating companies. The proposed team-project concept gained popularity, and many companies were anxious to support the students, who were soon competing for the most challenging projects. Word of the financial impact derived from the student project recommendations enhanced the competition among sponsoring companies to stay involved.
In 1999, Jim Bean left TMI to serve in two associate dean positions in the College of Engineering. Then, in 2004, he left UM for a position as dean at the University of Oregon, where he helped implement a similar team-focused program in their Sports Product Management Program. Today, Bean is provost and senior vice president for academic affairs at Northeastern University in Boston.
6.5.2. Later History (1999–2005)
Replacing Bean as the TMI engineering codirector was IOE professor Yavuz Bozer, who served the institute in that position until the fall of 2010. Bozer faced many challenges when taking this position. The very definition of manufacturing was changing, and some were claiming that manufacturing was dead in the United States. Labor-intensive work continued to be shipped overseas, and many plants in the United States were closing. Potential students were sensitive to these facts and expressed apprehension over joining a program with the word “manufacturing” in its very title.
Internally, the TMI codirectors worked with the admissions departments in the College of Engineering and the School of Business to inform students about the opportunities the institute presented. In order to continue on a successful path, The institute leadership doubled their efforts with the UM leadership to keep the institute’s programs viable and attractive as an option for the best and brightest students. At this time the public interest in supply chain management, operations management, and manufacturing engineering in general began to rise.
One issue was to achieve agreement by the faculty and industry sponsors that a change was needed in the institute’s name. Agreeing on the new focus and name was a daunting task. Multiple stakeholders were invited to suggest a new name. Alumni, prospective and current students, advisory boards, deans and faculty were all consulted. Finally, a new name was coined: Tauber Institute for Global Leadership (TIGL). (One year after the institute changed its name to reflect a global focus, MIT’s Leaders for Manufacturing became Leaders for Global Operations.)
Project sponsors continued to evolve, and the team projects gained additional recognition as valuable endeavors for companies. To further the teams’ contributions and learning and to smooth out any potential conflicts, a “team doctor” concept was initiated under Yavuz Bozer, which meant that in the rare event it was necessary, a staff member of the School of Business was deployed to help a team resolve occasional issues.
Bozer said that a well-known UM alumnus encapsulated Joel Tauber’s impact in one sentence: “As far as I know, TIGL makes the best use of its money, and I am very impressed with the remarkable impact you made with a relatively modest donation.” Funding, vision, and commitment have made TIGL an outstanding educational program for the College of Engineering and the School of Business. (For more information see the institute’s website: www.tauber.umich.edu/.)
6.6. The Financial Engineering Master’s Degree Program
The interdisciplinary Financial Engineering Program, offering a master’s of science in engineering, was created in July 1996 under the leadership of John R. Birge, who was then the chair of IOE. The original participating departments were IOE, from the College of Engineering, and mathematics and statistics, from the College of Literature, Science, and the Arts. Although not part of the program initially, the Ross School of Business allowed financial engineering students to enroll in the required business courses, and in May 2002 it became a full partner in the program. By 2005, other participating departments included EECS (from the College of Engineering), economics (College of Literature, Science, and the Arts), finance (Ross School of Business), and the multidisciplinary Center for Complex Systems.
Birge became the first director of the Financial Engineering Program, and IOE participating faculty included Vadim Linetsky and Jussi Keppo. After Birge left for Northwestern, subsequent directors from IOE included Romesh Saigal, Larry Seiford, and Stephen Pollock.
The program started with an initial enrollment of five students, but by the fall of 2005, it had an entering class of around 60 students and had grown from a 30-credit program that could be completed in three semesters to a 36-credit program requiring a summer and three semesters. Many financial engineering students co-registered in PhD programs in such departments as economics, mathematics, and finance. Most had previous experiences in the financial industry.
By 2005, graduates had been hired by investment banks (e.g., Morgan Stanley), corporate treasuries (e.g., Ford Credit), financial consulting firms (e.g., Bain), software providers (e.g., Infinity), energy companies, (e.g., Detroit Edison), banks (Credit Suisse), mutual funds (T. Rowe Price), hedge funds (Susquehanna Partners), government (World Bank), and consulting companies (Author Anderson).
One mechanical engineering/IOE student in the program won the Eurobanco Commercial Bank competition by increasing a paper portfolio from $250,000 to $12 million in three months.
6.7. The Hospital Systems Research Program
In the early 1960s, the department increased its interest in improving hospital operations. This effort was lead by Clyde Johnson and Dean Wilson, the latter of whom was also serving as the director of the Industrial Systems Laboratory within the department. Johnson and Wilson collaborated with the University of Michigan Hospital to establish the first IE group within a hospital to improve its operations. Before long this approach had spread to other hospitals in Michigan, Indiana, and Ohio. Johnson provided the strategic guidance and coordinated the various projects. Wilson supervised the students in determining and documenting the time it took to perform many of the repetitive tasks that occur in hospitals. The intent was to provide a more precise method of determining the staffing levels needed in different departments and under various operating conditions. Numerous time studies were performed. Essentially, this massive effort was analogous to the way a manufacturing operation plans and determines its productivity goals.
Many IE students worked on projects in these hospitals, and full-time IE staff positions became available to IE alumni. At first many of the projects focused on reducing cost and waste in support functions, such as the laundry, housekeeping, and food services departments. By 1963, the demand for IE projects was sufficient to spin off a nonprofit company: the Community Systems Foundation. Wilson, along with a recent IE PhD graduate, Bart Burkhalter, and a recent IE undergraduate, Karl Bartsch, were the initial officers. Within a short period they had consulting projects in almost all the local hospitals and were providing a large number of IE undergraduates and master’s students with excellent problems and experiences. The historical development of the foundation is described at its website: http://www.communitysystemsfoundation.org/downloads/CSF_50_Year_History.pdf.
During this period both Wilson and Johnson continued to teach in the IE Department, and in 1965, Richard Jelinek joined them as an assistant professor. Jelinek had just completed his PhD with Johnson as his adviser. Karl Bartsch left Community Systems Foundation and returned to the department to complete his master’s degree. He then formed Chi Systems, a second spin-off, which provided additional hospital consulting services. Wilson resigned his faculty position in 1968 to devote himself to the rapidly growing Community Systems Foundation. About this time the department’s Industrial Systems Laboratory, which had been providing staff support for the faculty and students to work in area hospitals, shut down because of political pressure from the now established private consulting groups, some of which were employing graduates of the department. In 1971, Jelinek moved to the faculty in the Department of Hospital Administration within the UM School of Public Health. Several years later he resigned from his faculty position to develop a new hospital systems consulting company in California. In 1974, Clyde Johnson retired, but his vision—along with his management of the many hospital projects performed by IE students and staff over his17 years with the department—had left a mark. The opportunities and means for all hospitals to improve their operations were now a matter of record, and companies like Community Systems Foundation and Chi Systems continued to expand their services throughout the ’70s and ’80s based on the earlier and continuing research being done in the department.
In 1970, Walton Hancock, who was chair of the IE Department and director of the Human Performance Laboratory, when studying ways to minimize hospital costs, became interested in whether the flow of patients was optimal. In order to learn more about patient flows and hospital management, he moved his primary office to the School of Public Health, where there was a hospital administration graduate program. There he joined John Griffin, the director of the program, and others in the School of Public Health to develop stochastic flow systems that achieved maximum occupancy once the number of beds to be staffed were properly determined. In 1974, one of Hancock’s first IOE PhD students working in hospital systems engineering, James Martin, joined Griffin and Hancock in the School of Public Health as an assistant professor. In 1976, Griffin, Hancock, and Fred Munson published the first book on the topic, Cost Control in Hospitals, which was lauded for its use of case studies showing that a data-driven systems approach could effectively reduce the cost of many different types of hospital operations. With the advent of mini computers in the ’70s, and with funding from a National Institutes of Health (NIH) grant, they developed a computer simulation meant to allow hospitals to maintain much higher occupancies, and at the same time provide for unexpected emergency arrivals. The first successful implementation of this simulation resulted in a six percent increase in patient occupancy with an increase in budget of only $10,000 annually in a large hospital. This was a time when NIH officials were promoting translational research. The project director of NIH encouraged Hancock to start a company to implement the Admission Scheduling and Control System (ASCS) in other hospitals. Hancock successfully implemented ASCS in 20 hospitals. In 1983, Hancock and Paul Walter published a widely acclaimed technical manual describing the process: The ASCS: Inpatient Admissions Scheduling and Control System. By this time Hancock had supervised seven IOE PhD students who contributed to a body of research on the use of various types of statistical and operations research methods to reduce hospital costs and to improve occupancy rates and patient care. One of their contributions was a computer-aided operating room scheduling system that enabled hospitals to increase utilizations by 20 percent and to start surgical procedures on time 95 percent of the time. This was implemented in six hospitals. Also, nurse staffing and scheduling algorithms were developed that allocated patients to nurses so that the probability that the nurse could do all of the work required was very high. It was estimated that implementation of ASCS would result in a budget reduction of over 30 percent in the more than 4,000 secondary hospitals in the country, and at the same time result in much higher quality of care. By 1995, a total of 12 students received their PhDs from this effort, nine of whom were IOE students.
In 1984, Hancock shifted his primary work back to the College of Engineering when he became the associate dean for the manufacturing initiative, though he still supervised several PhD students in the health care area. His impact on manufacturing operations led to his becoming the first William Clay Ford Chair of Product Manufacturing in 1989. With the end of Hancock’s leadership in healthcare systems during the ’80s there was a period when the department continued to offer a course in hospital systems and encouraged many undergraduate students to engage in projects in the UM Health System, but the department did not lead any related research. For the next 15 years the hospital systems course and student projects were mostly taught and supervised by adjunct professor Richard Coffey, a PhD graduate of the IOE Department in 1974. Coffey was the director of industrial engineering in the UM Health Systems. Like many important and complex problem areas, however, interest in the topic of health care did not disappear within the department, as discussed in chapter 8.
