Partitioning and balancing for the assembly of nozzle guide vanes in gas turbine engines

The gas turbine engine is used to power many different types of commercial and military aircraft. During the scheduled maintenance of these engines, many of the turbine components are replaced. Of particular importance to us is the replacement of nozzle guide vanes in the nozzle assembly section of the engine. Individual vanes are selected from inventory to make up sets, and each set must meet certain characteristics in order to be feasible. The vanes in each set must then be sequenced in order to meet additional criteria. In this article, we give heuristics for the above partitioning and sequencing problems. Empirical analyses, using actual data from a branch of the armed services and a major engine manufacturer, are used to evaluate the proposed heuristics. The results of these analyses indicate that the heuristics are effective.     

Optimizability of utility and value functions

For more than a decade, multiattribute utility/value theory and multiobjective mathematical programming have offered different approaches to similar problems. Unfortunately, the two areas have developed with little interaction in spite of their common aims. We consider the use of utility/value functions in a mathematical programming framework, and demonstrate that these functions often possess desirable properties from an optimization point of view. We conclude that a hybridization of approaches is more viable than is perhaps commonly assumed.     

The frequency assignment problem: A solution via nonlinear programming

This paper gives a mathematical programming model for the problem of assigning frequencies to nodes in a communications network. The objective is to select a frequency assignment which minimizes both cochannel and adjacent-channel interference. In addition, a design engineer has the option to designate key links in which the avoidance of jamming due to self interference is given a higher priority. The model has a nonconvex quadratic objective function, generalized upper-bounding constraints, and binary decision variables. We developed a special heuristic algorithm and software for this model and tested it on five test problems which were modifications of a real-world problem. Even though most of the test problems had over 600 binary variables, we were able to obtain a near optimum in less than 12 seconds of CPU time on a CDC Cyber-875.     

The complementary use of regression diagnostics and robust estimators

Regression modeling for prediction or forecasting purposes is critically dependent on the quality of the data which are used to estimate the model parameters. Extreme response or predictor-variable values can substantially influence least-squares estimates and disproportionately affect predictions. Robust alternatives to least-squares are less sensitive to extreme observations and can provide more precise predictions. In this article diagnostic displays are used to identify extreme observations and to assess the sensitivity of least-squares parameter estimates and predictions to the inclusion of these observations in a data set. The displays are shown to aid in the interpretation of weights which robust estimators assign to influential observations.     

British Policy on Total Maritime Warfare and the Anti-Shipping Campaign in the Mediterranean, 1940–1944

During World War II, the British ran a sustained anti-shipping campaign against Axis merchant and supply traffic in the Mediterranean. Although the effects of this on the land war in North Africa have been the subject of much debate, little attention has been paid to the nature and prosecution of the campaign itself. This article analyses the changes in British attitudes and policy towards attacking merchant shipping prior to and throughout the campaign. It then goes on to examine the conduct of the campaign itself and compare it with other British efforts elsewhere during the war. It concludes that the anti-shipping campaign in the Mediterranean was a unique combined arms offensive for the British, and a major evolution in their attitudes and policy towards maritime total war.     

Rent‐seeking models of international competition: An experimental investigation*

This paper experimentally tests a number of hypotheses that follow from models of international competition that are based on Tullock style rent‐seeking models. Specifically, we designed and performed experiments to see how variations in the degree of publicness in the prize as well as changes in the values assigned to the prize affect alliances in terms of individual nation contributions, total contributions, burden sharing, and the likelihood of winning the prize. While there was substantial variation in individual behavior, the results of the experiments generally conform to the theoretical predictions of the rent‐seeking model of international competition and alliances.