Executive summaries of IIE Transactions and The Engineering Economist
Edited by Susan Albin and Joseph C. Hartman
This month our research highlights focus on reliability and warrantees. The first summary presents a method to determine whether it is worthwhile, and if so how much, to upgrade secondhand products before they are sold and the warranty contract begins. The second summary estimates system reliability by allocating testing resources to various component types. These articles will appear in the January issue of IIE Transactions (Volume 44, No. 1).
Fast, online measurements of wafers in semiconductor manufacturing
A wafer is a thin slice of semiconductor material, such as silicon, used in the fabrication of integrated circuits. Micro-electronic devices are grown on the wafer in a multiple-step process. It is ideal if the wafer is perfectly flat and uniformly thick with no warping. Flatness, thickness and warp are measured with sensors. The results are termed geometric profiles.
Fast and accurate measurement of geometric profiles would help product quality assurance, process monitoring and quality improvement. However, current wafer profile measurement is time-consuming. For example, it takes more than eight hours to measure the profiles of 400 wafers. This has made wide application of the geometric profile measurements in semiconductor manufacturing processes impossible. Currently, these measurements only can be made in offline processes.
This research is motivated by the need for a fast, online measurement strategy for wafer geometric profile estimation. In the article “Sequential Measurement Strategy for Wafer Geometric Profile Estimation,” assistant professor Ran Jin of Virginia Tech and doctoral student Chia-Jung Chang and professor Jianjun Shi from Georgia Tech propose a sequential sensing strategy that yields satisfactory accuracy and significantly reduces measurement time.
In the proposed approach, initial samples are measured first. Then a process model is fitted to estimate the true profile of a wafer. The profile prediction and its uncertainty serve as guidelines to determine the measurement locations for the next sampling iteration. A case study is provided to illustrate the procedures and effectiveness of the proposed methods based on the wafer thickness profile measurement in slicing processes.
The research is partly supported by the National Science Foundation-sponsored project “NSF – 1030125: Metamodel-based Measurement, Control, and Optimization of Engineered Surfaces.”
CONTACT: Jianjun “Jan” Shi; email@example.com; (404) 385-3488; H. Milton Stewart School of Industrial and Systems Engineering, Georgia Institute of Technology, 765 Ferst Drive NW, Atlanta, GA 30332-0205
Evaluate conflicting responses to reach the best design sooner
In designing processes, it is critical to determine settings for input variables that optimize the main quality characteristic value, called the response. In practice, it is common to have multiple responses of interest. And often, these multiple responses conflict in that optimizing one of them may lead to a poor result for another response. To obtain a satisfactory compromise, it is important that the decision maker’s view of the importance of each of the responses is incorporated into the solution process.
The work here was motivated by a problem that the authors encountered in the steel industry. The mechanical properties of hot rolled steel strips are measured by three different responses: tensile strength, yield strength and elongation. Each response has a different unit of measurement. There is an additional complicating factor that the three responses (especially the first and third responses) tend to increase or decrease together because they are correlated. Thus, it is not straightforward to obtain the desired balance among the three responses.
In “An Interactive Method to Multiresponse Surface Optimization Based on Pairwise Comparisons,” Dong-Hee Lee, a senior engineer at Samsung Electronics and a former Ph.D. student at Pohang University of Science and Technology (POSTECH), professor Kwang-Jae Kim of POSTECH and professor Murat Köksalan of Middle East Technical University develop a new interactive method that accounts for multiple responses when optimizing input settings.
In contrast to many existing methods, their method does not require the decision maker to give response importance information before solving the problem. Instead, it extracts the information progressively in an interactive manner by asking the decision maker to choose between selected pairs of designs, a task that can be performed reliably. The authors demonstrate that their method obtains satisfactory compromise solu-tions with just a few pairwise comparisons.
CONTACT: Kwang-Jae Kim; firstname.lastname@example.org; +82-54-279-2208; Department of Industrial and Management Engineering, Pohang University of Science and Technology (POSTECH), Pohang, Kyungbuk 790-784, Republic of Korea
The most recent issue of The Engineering Economist (Volume 56, Number 3) is incredibly diverse. The initial article reviews publishing trends in the field of engineering economics and includes insight into today’s most relevant issues. The article notes that incorporating risk into analysis has been the most prevalent topic for the past decade, with real options analysis leading the way. A second article outlines better methods in which to generate scenarios in stochastic programming, and another article outlines what should be taught regarding engineering economy in undergraduate engineering programs. The two articles below highlight the use of real options in making investment decisions.
Managing with real options
Real options were designed in order to allow the incorporation of management flexibility into the value of an investment.
Traditional net present value (NPV) analysis requires the estimation of an investment project’s cash flows, which are used then to compute its value with an appropriate interest rate. Proponents of real options claim that this method undervalues opportunities as changes in the future can enhance the value of a project. That is, NPV analysis undervalues the worth of a project because management has flexibility in its execution.
In real option analysis, the worth of the project is computed by taking the traditional NPV value and adding the value of the option for the project.
While this approach has merit and validity, one underlying assumption is that management can take advantage of any presented flexibility in the project. If not, then the value of the project is being overestimated. This paper attempts to close this gap in practice and analysis by actively managing the options. Specifically, in “Active Management of Real Options,” professor Tom W. Miller of Kennesaw State University models business opportunities to introduce new products at a later date as real options. If the options are not worth exercising (flexibility in the project does not enhance it enough to invest at the current moment), then the manager can invest (if deemed prudent) during the option’s life to move it closer to being profitable.
In technical terms, the “distance to breakeven” is the measure used to guide management. The distance from breakeven measures how far, in standard deviations, the opportunity is expected to be from being worthy of investment by its expiration date (when the option to invest is no longer feasible). The distance to breakeven and the value of the business opportunity may be managed actively during the real option’s life by invest-ment in advertising and research and development activities. Product design and process design changes may be used to improve the value of the investment and its options.
A case study shows that active manage-ment can create substantial positive incremental value for good or marginal opportunities. That is, if the analysis shows that a project does not have far to “move” in terms of distance to breakeven, then management should invest in advertising or research and development. However, these activities should not be pursued for bad investments – those with big distances
CONTACT: Tom W. Miller; (770) 423-6494; tmiller@.kennesaw.edu; Department of Economics, Finance and Quantitative Analysis, Coles College of Business, Burruss Building 348, Kennesaw State University, Kennesaw, GA 30144
Using cash flow estimates in real options analysis
It has been reported that the adoption of real options in practice has been limited. Numerous reasons have been noted, including that the approach is too complicated and that the necessary data is not available. There also are great discrepancies among academics and practitioners as to how the volatility should be estimated for real options analysis.
In the paper “Integrating Real Options with Managerial Cash Flow Estimates,” Kelsey Barton and professor Yuri Lawryshyn of the University of Toronto attempt to overcome a number of these shortcomings. First, managers are expected to provide optimistic, likely and pessimistic estimates of the cash flows for a project over its life. (This is typical in practice for many estimates.) Then, these estimates are correlated to a relevant, traded index such that a volatility measure is defined and the real option can be valued. If the real option exhibits some correlation to the market, then a hedging strategy can mitigate some of the market exposure by selling shares in the correlated index.
Interestingly, traditional real option theory has shown that the value of the investment increases with the volatility of the investment because it drives up the value of the option associated with the investment. This new approach reveals that the value of the real option is not necessarily an increasing function of volatility as the real option value is contingent on the present value of the future cash flows, a value capture by risk-neutral expectations.
The approach truly has the ability to further the application of real options in practice. Instead of relying on future cash flow estimates and simulation, the volatility measure – arguably the most controversial estimate required in real options analysis – can be derived from good, average and poor outcome estimates. It may lead to further adoption. The authors illustrate their method on an investment in unmanned aerial vehicles.
CONTACT: Yuri Lawryshyn; (416) 946-0576; email@example.com; Department of Chemical Engineering and Applied Chemistry, University of Toronto, 200 College St., Toronto, Ontario, M5S 3E5 Canada
Susan Albin is a professor at Rutgers University in the Department of Industrial and Systems Engineering. She is editor-in-chief of IIE Transactions and a fellow of IIE.
Joseph C. Hartman is editor of The Engineering Economist. He is a professor and the department chair of industrial and systems engineering at the University of Florida. He has been a member of IIE since 1995 and previously served on the IIE Board of Trustees as senior vice president for publications.
IIE Transactions is IIE’s flagship research journal and is published monthly. It aims to foster exchange among researchers and practitioners in the industrial engineering community by publishing papers that are grounded in science and mathematics and motivated by engineering applications.
The Engineering Economist is a quarterly refereed journal devoted to issues of capital investment. Topics include economic decision analysis, capital investment analysis, research and development decisions, cost estimating and accounting, and public policy analysis.
To subscribe, call (800) 494-0460 or (770) 449-0460.