Knowledge that Transforms

An individual is said to be potentially miscalibrated if he is not sure whether his future subjective probability assessments will agree with observed frequency. Alternately, the individual is said to be uncertain about his own calibration. It is argued that such a person will never perceive any two events as (probabilistically) independent, in the same sense that an ignorant person does not perceive events as certain. Uncertainty about one's own calibration does not prevent rational behavior in the decision theoretic sense, but it may make much more difficult the process of translating decision theoretic principles into practical procedures for analysis of real decision problems.

A quantitative comparison between three types of capacity planning or shop loading techniques is presented. The results were obtained from simulation experiments.

The problem of optimal planning of an electric power system is formulated as a cost minimizing mixed-integer mathematical program. The model emphasizes system reliability, requiring reserve generating capacity such that a Loss of Load Probability criterion of one day in ten years is met. Integer variables are used in the model to express system reliability explicitly as a constraint on the system model. Results are given for several plant outage rates and demand levels. The results show that the reliability and “rule-of-thumb” approaches to system reserve planning can yield substantially different optimal system costs and configurations.

Because R&D and marketing are dependent upon each other for new product development, it is imperative that they achieve consensus and organizational integration (a team spirit of collaboration and joint commitment). But, consensus and integration are often inhibited by the differing viewpoints of R&D and marketing, which are a natural consequence of their specialized organizational roles and cultures. There is a need for a process that will bridge these dissonant viewpoints and cultures, while otherwise preserving the specialized orientations of the two parties. The bridging properties of three group decision making processes—nominal, interacting, and combined nominal-interacting—were tested by nine strategic planning teams, each composed of R&D and marketing personnel. The combined nominal-interacting process yielded very high levels of statistical consensus and group integration. The nominal process produced statistical consensus but it did not yield high levels of integration. The interacting process did not produce either consensus or integration. The results here and elsewhere indicate that consensus and collaboration problems between R&D and marketing may be alleviated by replacing the interacting decision making process, which is typically used by many organizations, with a combined nominal-interacting process.

We consider the problem of ordering perishable inventory when there is uncertainty in both the demand and the lifetime of the product. Under the assumption that units outdate in the same order in which they enter inventory, it is shown that the structure of the optimal policy is essentially the same as in the case where the lifetime is deterministic. An explicit expression for the expected outdating of any order is derived. Two different bounds on the expected outdating are then used to construct two critical number approximations. Computations for a discrete version of the problem are performed to compare the expected costs of both approximations with the optimal. One approximation appeared to give slightly better results and produced an expected cost generally within a fraction of a percent of the optimal for the cases tested.

We point out some errors in Iglehart's paper, “The Dynamic Inventory Problem with Unknown Demand Distribution,” published in Management Science in 1964.

The problem of choosing the one best or several best of a set of sequentially observed random variables has been treated by many authors. For example, the seller of a house has this problem when deciding which bids on the house to accept and which to reject. We assume that the bids are identically distributed random variables and at most n can be observed. Each bid is accepted or rejected when received; a bid rejected now cannot be accepted later on. The object is to maximize the expected value of the bid actually accepted. Unlike most previous authors, we examine the case where one or more parameters of the common underlying distribution are unknown and information on these is updated in a Bayesian manner as the successive random variables are observed. Using the properties of location and scale parameters, an explicit form for the optimal policy is found when the underlying distribution is normal, uniform, or gamma and the prior is from the natural conjugate family. Simulation results concerning sensitivity of the value obtained to the amount and correctness of the prior information for these three families is then presented.

This study focuses attention on methods for generating useful solution standards for large combinatorial problems. In particular, several procedures that provide point estimates of the value of the optimum solution are suggested and tested. These concepts are applied to a representative combinatorial problem: flow shop sequencing. Detailed computational results are presented.