Interfaces

In an era of limited healthcare budgets, mathematical models can be useful tools to identify cost-effective programs and to support policymakers in informed decision making. This paper reports results of our work carried out over several years with the Asian Liver Center at Stanford University, a nonprofit outreach and advocacy organization that is an international leader in the fight against hepatitis B and liver cancer. Hepatitis B is a vaccine-preventable viral disease that, if untreated, can lead to death from cirrhosis and liver cancer. Infection with hepatitis B is a major public health problem, particularly in Asian populations. We used new combinations of decision analysis and Markov models to analyze the cost-effectiveness of several interventions to combat hepatitis B in the United States and China. The results of our OR-based analyses have helped change United States public health policy on hepatitis B screening for millions of people and have helped encourage policymakers in China to enact legislation to provide free catch-up vaccination for hundreds of millions of children. These policies are an important step in eliminating health disparities, reducing discrimination, and ensuring that millions of people who need it can now receive hepatitis B vaccination or lifesaving treatment.

We tested the effects of various of policy rules on retailer behavior in laboratory experiments conducted at Hewlett-Packard Laboratories. Our experimental design models the multifaceted contemporary market for consumer computer products and is quite complex, but we found that participants can make effective decisions and that their behavior is sensitive to variations in policies. Based on our results, Hewlett-Packard changed its policies; for example, it made the consequences for violations forward-looking as well as backward-looking). This line of research appears promising for complex industrial environments.

Recent research shows that journal reviewing practices are neither objective nor fair. I propose a procedure to increase the likelihood of publishing important papers. This will be tested by Interfaces for a year.

I designed a voting system for a professional association to ensure the equitable representation of different interests on its governing board. Approval voting, whereby voters can vote for as many candidates as they approve of, is combined with constraints on the numbers that can be elected from different categories of members to find the set of candidates most approved of by all voters, subject to the constraints.

Policymakers need to know whether prediction is possible and, if so, whether any proposed forecasting method will provide forecasts that are substantially more accurate than those from the relevant benchmark method. An inspection of global temperature data suggests that temperature is subject to irregular variations on all relevant time scales, and that variations during the late 1900s were not unusual. In such a situation, a "no change" extrapolation is an appropriate benchmark forecasting method. We used the UK Met Office Hadley Centre's annual average thermometer data from 1850 through 2007 to examine the performance of the benchmark method. The accuracy of forecasts from the benchmark is such that even perfect forecasts would be unlikely to help policymakers. For example, mean absolute errors for the 20- and 50-year horizons were 0.18 � oC and 0.24 � oC respectively. We nevertheless demonstrate the use of benchmarking with the example of the Intergovernmental Panel on Climate Change's 1992 linear projection of long-term warming at a rate of 0.03 � oC per year. The small sample of errors from ex ante projections at 0.03 � oC per year for 1992 through 2008 was practically indistinguishable from the benchmark errors. Validation for long-term forecasting, however, requires a much longer horizon. Again using the IPCC warming rate for our demonstration, we projected the rate successively over a period analogous to that envisaged in their scenario of exponential CO2 growth--the years 1851 to 1975. The errors from the projections were more than seven times greater than the errors from the benchmark method. Relative errors were larger for longer forecast horizons. Our validation exercise illustrates the importance of determining whether it is possible to obtain forecasts that are more useful than those from a simple benchmark before making expensive policy decisions.

Back haul problems occur in many areas of transportation. One-way rental often takes equipment, such as cases and containers, from an area of high demand to an area of low demand. The problem is to return the equipment to the location of need, a problem typically viewed as an administrative and scheduling problem. We developed a decentrali7cd approach in which a specially designed market organizes competition and information to minimize the cost of back-hauls without the direct intervention of administrative negotiations or command-and- control types of scheduling. We employed laboratory experimental methods to test the concept, examine its performance against theoretical benchmarks, and explore its limitations.

It is often claimed that managers do not read serious research papers in journals. If true, this neglect would seem to pose a problem because journals are the dominant source of knowledge in management science. By examining results from the forecasting principles project, which was designed to summarize all useful knowledge in forecasting, we found that journals have provided 89 percent of the useful knowledge. However, journal papers relevant to practice are difficult to find because fewer than three percent of papers on forecasting contain useful findings. That turns out to be about one useful paper per month over the last half-century. Once found, the papers are difficult to interpret. Managers need low-cost, easily accessible sources that summarize advice (principles) from research; journals do not meet this need. To increase the rate of progress in developing and communicating principles, researchers, journal editors, textbook writers, software developers, web site designers, and practitioners should make some changes. Some examples: Researchers should directly study forecasting principles. Journal editors should actively solicit papers – invited submissions were about 20 times better than standard submissions at producing useful findings that were often cited, and does so at a lower cost. Web-site and software developers should provide practitioners with low-cost ways to use principles. Practitioners should apply the principles that are currently available.

Spicer Off-Highway Products Division (a division of Dana Corporation) asked us to develop OR tools to improve the due date performance and to shorten the manufacturing lead time of its powershift transmission plant in Brugge, Belgium. We modelled the manufacturing system as a queueing model, and used the model to analyze and evaluate improvement schemes (layout changes, product-mix decisions, lot-sizing decisions, and lead time estimations). Next, we developed a finite scheduler to improve the detailed scheduling of the shop. Our modelling effort contributed to the successful combination of analysis, planning and detailed scheduling. The plant increased productivity by 27.3 percentage points, decreased manufacturing lead times by a factor two to three, increased its workforce by 41 percent, and decreased its operating costs. The division is now profitable.

We examined the relationships between the research originating at business schools, students^R satisfaction with the schools, and the published ratings of the school^Rs prestige. Research was positively correlated to prestige (where prestige was based on the perceptions of academics, firms, and student candidates). The satisfaction of recent graduates was not related to a school^Rs prestige (based on the perceptions of academics and business firms). Research productivity of schools was not associated with lower satisfaction among their recent graduates. We conclude that schools should emphasize research instead of teaching if they desire high prestige.

A combinatorial auction is an auction where multiple items are for sale simultaneously to a set of buyers. Furthermore, buyers are allowed to place bids on subsets of the available items. This paper focusses on a combinatorial auction where a bidder can express his preferences by means of a so-called ordered matrix bid. This matrix bid auction was developed by Day (2004) and allows bids on all possible subsets, although there are restrictions on what a bidder can bid for these sets. We give an overview of how this auction works. We elaborate on the relevance of the matrix bid auction and we develop methods to verify whether a given matrix bid satisfies a number of properties related to micro-economic theory. Finally, we investigate how a collection of arbitrary bids can be represented as a matrix bid.Keywords: Combinatorial auction, matrix bids, free disposal,subadditivity, submodularity, gross substitutes, expressiveness

Ever more companies are recognizing the benefits of closed-loop supply chains that integrate product returns into business operations. IBM has been among the pioneers seeking to unlock the value dormant in these resources. We report on a project exploiting product returns as a source of spare parts. Key decisions include the choice of recovery opportunities to use, the channel design, and the coordination of alternative supply sources. We developed an analytic inventory control model and a simulation model to address these issues. Our results show that procurement cost savings largely outweigh reverse logistics costs and that information management is key to an efficient solution. Our recommendations provide a basis for significantly expanding the usage of the novel parts supply source, which allows for cutting procurement costs.

This paper examines the introduction of information-sharing into the supply chains for pharmaceutical products in the United States. This introduction was unusual for several reasons. First, it was catalyzed from outside the industry, by a Securities and Exchange Commission (SEC) investigation into improper financial reporting by a single manufacturer. Second, it was initiated by pharmaceutical manufacturers in order to keep distributor inventories low. Third, although its effect on pharmaceutical distributors has been profound, evidence indicates that information-sharing has had no impact on pharmaceutical manufacturers' own inventorymanagement practices.

Formal education can be improved by transferring responsibility from the teacher to the learner. A simple approach to this is the time contract. Time contracts have been used successfully in nine quasi-experiments but, despite these successes, some educators see this as subversive research.

The commentators raised many interesting ideas in response to Armstrong and Pagell (2003), from which one general theme emerges: The commentators claim that management science lacks the incentives to encourage efforts to develop and communicate grounded principles. As a result, academics often conduct their research as an intellectual exercise with little concern as to whether their findings might eventually be of any practical use. The problem extends beyond management science. Smith (1991), an editor of the British Medical Journal, concluded from a review that only about 15 percent of medical interventions are supported by solid scientific evidence. He attributes this disconnect to an estimate that only about one percent of articles in medical journals are scientifically sound. Such results indicate problems with incentives in research.

Sophisticated extrapolation techniques have had a negligible payoff for accuracy in forecasting. As a result, major changes are proposed for the allocation of the funds for future research on extrapolation. Meanwhile, simple methods and the combination of forecasts are recommended.

In important conflicts, people typically rely on experts' judgments to predict the decisions that adversaries will make. We compared the accuracy of 106 expert and 169 novice forecasts for eight real conflicts. The forecasts of experts using unaided judgment were little better than those of novices, and neither were much better than simply guessing. The forecasts of experts with more experience were no more accurate than those with less. Speculating that consideration of the relative frequency of decisions might improve accuracy, we obtained 89 forecasts from novices instructed to assume there were 100 similar situations and to ascribe frequencies to decisions. Their forecasts were no more accurate than 96 forecasts from novices asked to pick the most likely decision. We conclude that expert judgment should not be used for predicting decisions that people will make in conflicts. Their use might lead decision makers to overlook other, more useful, approaches.

An approach is described to develop a formal corporate strategic planning process in a business firm. Three levels of planning are recognized: corporate, business, and functional. A methodology is presented suggesting the tasks that have to be performed at each level leading toward the formulation of strategies, programs, and budgets. Examples are offered to illustrate the applicability of this methodology.

Although to date computer technology has been used primarily to automate paperwork, recently we have entered an era in which information availability and new communication processes are having a significant impact on corporate life. In a conceptual model of this phenomenon, one aspect, the way in which technology is now driving corporate strategy, is emphasized.

I worked on modeling integrated steel plants in India (as an analyst) and in the USA (as a doctoral student). The following factors influence the successful OR/MS practice: positioning of OR/MS in Operations and not in Research and Development, selecting the right problem, emphasising on problem solving and not on model building, learning lessons from failure, right academic and practical training of OR teams, experiences of working in the shop-floor of a factory. Universities should give more emphasis in teaching OR as research in operations and to the process of implementation.

We ask various experts, who produce sales forecasts that can differ from earlier received model-based forecasts, what they do and why they do so. A questionnaire with a range of questions was completed by no less than forty-two such experts who are located in twenty different countries. We correlate the answers to these questions with actual behavior of the experts. Our main findings are that experts have a tendency to double count and to react strongly to recent volatility in sales data. Also, experts who feel more confident give forecasts that differ most from model-based forecasts.

Information technology and its driving cost performance imperative have changed traditional labor to capital relationships and opened up new opportunities for strategic use of the technology. Different forms of interconnection realized through improved interconnection and larger more accessible data bases are often the basis for deriving strategic advantage. A search strategy based on electronic integration can help organizations find potentially strategic applications.

Although the past 20 years seem to indicate that modeling for public policymaking has not been very successful, as better decision support systems and other user-friendly computer systems are developed, analysts, policymakers, and staff members will have an opportunity to work more closely throughout the whole process from problem definition to implementation. Policy analysts will be in greater demand in government and as consultants to government agencies.

The number of on-line computer applications that exist today gives ample testimony that the computer hardware business has come of age. About fifteen years ago, IBM built the SAGE system which, for the first time, tied computers together with a number of other components. A large central processor communicated with a number of peripheral processors that controlled radar units. Every component had to be specially designed and built at no small cost. Today, the components of a similar system would be standard, with a computer, the interfaces, and the teleprocessing all catalog items.

Before 1960, little empirical research was done on forecasting methods. Since then, the literature has grown rapidly, especially in the area of judgmental forecasting. This research supports and adds to the forecasting guidelines proposed before 1960, such as the value of combining forecasts. New findings have led to significant gains in our ability to forecast and to help people to use forecasts. What have we reamed about forecasting over the past quarter century? Does recent research provide guidance for making more accurate forecasts, obtaining better assessments of uncertainty, or gaining acceptance of our forecasts?

Combined-value auctions (CVAs) allow participants to make an offer of a single amount for a collection of items. These auctions provide value to both buyers and sellers of goods or services in a number of environments, but they have rarely been implemented, perhaps because of lack of knowledge and experience. Sears Logistics Services (SLS) is the first procurer of trucking services to use a CVA to reduce its costs. In 1993, it saved 13 percent over past procurement practices. Experimental economics played a crucial role in the development, sale, and use of the CVA.