EXPERIMENTAL RESULTS OF AN APPROACH FOR EVALUATING MULTIATTRIBUTE ALTERNATIVES

In this paper the results of an extensive simulation study are reported to provide empirical evidence on the Evaluation and Bound approach. Some rules for implementing the procedure interactively on a computer are also examined and compared experimentally. The results suggest that the Evaluation and Bound procedure, in conjunction with an appropriate strategy of implementing it, would reduce considerably the information requirement of the decision maker in the choice of a preferred multiattribute alternative.     

On estimating the conditional probability of discovering a new species

Based on a capture-recapture sample of size $i;n+k,n≥ l,k ≥ 0, from a population of an unknown number of distinct species (or classes), the problem of estimating the total probability of the species unobserved in the first n selections is considered. As the estimand depends on both the unknown parameters and the data, the standard theory of estimation is inadequate for this problem A suitable definition of sufficiency is introduced and used to prove a Rao-Blackwell type result and discuss uniformly minimum mean squared error unbiased estimation. An alternative proof for an inadmissibility result is presented. The new proof gives more insight and a method for deriving improved estimators. The theoretical developments may be useful in other problems concerning inferences about random parametric functions.     

Scheduling to minimize stringer utilization for continuous annealing operations

In this paper we present a mathematical program and heuristic algorithms to schedule coils for the production operations in a copper (or steel) coil manufacturing industry. The processing facility uses continuous operations for processing (e.g., galvanizing and annealing) while the handling unit is a discrete coil. The ends of coils are “stitched” or welded together to enable continuous processing, and the joint is later sheared off to obtain the processed coils. Processing constraints impose restrictions on the compatibility between a pair of coils that are overcome by introducing a dummy coil called stringer, which is very expensive to a mill. This paper deals with modeling the sequencing/scheduling problem of coils on parallel non-identical machines to minimize stringer utilization. Both computational and practical experiences show the efficiency and effectiveness of the solution approaches. Implementing these methods in an actual coil annealing facility resulted in 65% reduction in stringer utilization.     

International labor migration and domestic labor supply

Prominent among the variety of issues raised by the immigration of labor are its effects on domestic wage rates and labor supply. In contrast to the existing literature, this paper constructs a dynamic, general equilibrium framework to study the relationship between international labor migration and domestic labor supply. The general equilibrium nature of the model enables us to endogenize the pattern of labor migration. The effect of labor migration on domestic wage rates and labor supply is shown to depend on the pattern of labor migration. If the substitution effect dominates the income effect in labor supply, the domestic supply of labor necessarily decreases in response to an inflow of migrants. This happens even if immigrants, through their savings behavior, cause an increase in the domestic capital-labor ratio and wage rate. Similarly, if the dominant effect is the income effect, the immigration of labor necessarily increases the domestic supply of labor.I wish to thank Oded Galor, Harl Ryder, Kalpana Kochhar and two referees for valuable comments. The opinions expressed in this paper are my own and do not necessarily reflect the views of Joel Popkin and Company.     

Evaluating Time Streams of Income: Discounting What?

For decisions whose consequences accrue over time, there are several possible techniques to compute total utility. One is to discount utilities of future consequences at some appropriate rate. The second is to discount per-period certainty equivalents. And the third is to compute net present values (NPVs) of various possible streams and to then apply the utility function to these net present values. We find that the best approach is to first compute NPVs of various possible income streams and then take the utility of such NPVs. We show the drawbacks of other alternative models of evaluating income streams. The article discusses the advantages of the power and logarithmic forms in the modeling of time preference. These are the only forms for which utility of income and utility of consumption are strategically equivalent. Further, these forms permit the flexibility in the choice of a time period (e.g., monthly or quarterly) without modifying the utility function, thus simplifying analysis.     

The impact of lean practices on operational performance – an empirical investigation of Indian process industries

In deciding to adopt lean manufacturing, it is imperative to investigate where and how lean practices are most needed to influence manufacturing and business performance. Such an investigation becomes indispensable when lean thinking is to be considered in a production arrangement different to the conventional, repetitive, high-volume, stable-demand and discrete-manufacturing environment. This study provides explanations of how performance is improved through the adoption of lean practices in process industries. This is a relatively under-researched area compared to the performance effects associated with the introduction and implementation of lean principles in traditional, discrete manufacturing. Based on a survey of Indian process industries, this study attempts to develop an empirical relationship between lean practices and performance improvement through the use of multivariate statistical analysis. The findings have led to the conclusion that lean practices are positively associated with timely deliveries, productivity, first-pass yield, elimination of waste, reduction in inventory, reduction in costs, reduction in defects and improved demand management. However, within a process-industry context, lean practices related to pull production were found to have a marginal impact on performance improvement. A detailed discussion of the findings along with their theoretical and managerial implications is provided in the paper.     

Admissibility of a test procedure in a conditionally specified random model

Admissibility of a test procedure is a desirable property though not a compelling one; while lack of it often results in discarding a given test procedure. In this paper we have proved necessary conditions for admissibility for a test procedure.     

Testing equality of conditionally jfclhwdqw exponential distributions

For the models given V = v (a common random stress), X and Y are independently exponentially distributed with failure rates λ₁and λ₂v, testing H₀λ₁λ₂using a random ‘paired’ sample is considered. It is shown that a uniformly most powerful invariant test does not exist even for one sided alternatives; locally most powerful invariant tests are derived and compared with existing procedures. The method is illustrated with reliability data. Finally, the robustness of the tests when the relationships of the failure rates to V is more complex are established.     

The concept of risk unbiasedness in statistical prediction

This paper extends the concept of risk unbiasedness for applying to statistical prediction and nonstandard inference problems, by formalizing the idea that a risk unbiased predictor should be at least as close to the “true” predictant as to any “wrong” predictant, on the average. A novel aspect of our approach is measuring closeness between a predicted value and the predictant by a regret function, derived suitably from the given loss function. The general concept is more relevant than mean unbiasedness, especially for asymmetric loss functions. For squared error loss, we present a method for deriving best (minimum risk) risk unbiased predictors when the regression function is linear in a function of the parameters. We derive a Rao–Blackwell type result for a class of loss functions that includes squared error and LINEX losses as special cases. For location-scale families, we prove that if a unique best risk unbiased predictor exists, then it is equivariant. The concepts and results are illustrated with several examples. One interesting finding is that in some problems a best unbiased predictor does not exist, but a best risk unbiased predictor can be obtained. Thus, risk unbiasedness can be a useful tool for selecting a predictor.