Improving perishable product inventory management using goal programming

The multiple objectives of perishable product inventory management are examined in this paper. These objectives include: (1) satisfying demand by carrying sufficient inventories, (2) holding down inventory carrying costs, (3) keeping the amount of product spoilage (outdating) at an acceptable level, and (4) maintaining quality by using the product while it is still fresh, and (5) keeping the cost of rotation low. Some of the above objectives are in conflict. Thus, certain redistribution policies may help attain one or more of these objectives to a greater extent. Redistribution involves the transfer of the product from outlets where demand is low to outlets where demand is high. A goal programming model for solving redistribution problems is presented. An example is provided and sensitivity analysis is performed in order to determine when redistribution is advantageous. Applicability of the model is discussed.     

Robustness to the unavailability of data in block designs

We present general results for finding or boundingt _maxB , the maximum number of arbitrary whole blocks of observations which can be removed from a block design, and still leave all of the elementary treatment contrasts estimable. The block sizes may be larger than the number of treatments. The results are applied to BBDs, reinforced BIBDs and BBDs, BTIBDs, and a series of variance balanced incomplete block designs with two block sizes. Also given for most of these designs, are results fort _max, the maximum number of arbitrary, scattered observations that can become unavailable, and still leave all of the elementary treatment contrasts estimable.The work was undertaken while Dr. Whittinghill was visiting Ohio State University, and supported by a grant from the Natural Sciences Division, Colby College, Waterville, Maine.     

Testing equality of variances in the analysis of repeated measurements

The problem of comparing the precisions of two instruments using repeated measurements can be cast as an extension of the Pitman-Morgan problem of testing equality of variances of a bivariate normal distribution. Hawkins (1981) decomposes the hypothesis of equal variances in this model into two subhypotheses for which simple tests exist. For the overall hypothesis he proposes to combine the tests of the subhypotheses using Fisher's method and empirically compares the component tests and their combination with the likelihood ratio test. In this paper an attempt is made to resolve some discrepancies and puzzling conclusions in Hawkins's study and to propose simple modifications.
The new tests are compared to the tests discussed by Hawkins and to each other both in terms of the finite sample power (estimated by Monte Carlo simulation) and theoretically in terms of asymptotic relative efficiencies.     

Monotonizing empirical Bayes estimators for a class of of discrete distributions with monotone likelihood ratio

A procedure is given for the construction of a monotone estimator that dominates a given estimator for a class of discrete distributions with monotone likelihood ratio. This procedure is applied to some empirical Bayes estimators. Monte Carlo results are given that demonstrate the usefulness of monotonizing.     

Aggregation effects on price and expenditure elasticities in a quadratic almost ideal demand system

Abstract.  While it is well known that demand elasticities calculated at the macro level will in general differ from those calculated at the micro level because of aggregation effects, there remain the questions of how large the effects are and how they vary with the degree of inequality in the income distribution. We explore these questions with models based on a quadratic version of the Almost Ideal Demand System. We investigate the elasticity differences theoretically and then calibrate the models and generate numerical results, using income data for seven countries with widely different distributions. The aggregation effects are found generally to be rather small, even with highly unequal income distributions.     

Forecasting first births

This study is concerned with one aspect of the family cycle, namely, the transition from young married to young married with small children. The focus is on developing models to forecast entries into this latter stage for the purpose of marketing research. "Using ordinary least squares, forecasting models were estimated for (1) total number of first births, (2) number of white first births, and (3) number of nonwhite first births." Models are estimated for both the United States and California using data from official sources.     

Forecasting and random processes

On the simple model of a (univariate) random process some general problems of “random logic” are discussed. New random approaches to explorative, normative and systematic (normex) forecasting are developed, and some applications are suggested.