Arthur J. Swersey

AN EVALUATION OF PRE-CONTROL

In this paper we discuss pre-control, an alternative to statistical control charts for monitoring processes. Pre-control and standard control charts are compared; their different objectives are emphasized and their drawbacks, as well as their advantages..

Dorian Shainin's Variables Search Procedure: A Critical Assessment

In this paper we discuss Variables Search, one of Dorian Shainins statistical tools for quality improvement. We are critical in our evaluation of Variables Search as an experimental design method for uncovering important factors. We find that there is ..

Chapter 6 The deployment of police, fire, and emergency medical units

Publisher Summary This chapter discusses the deployment of police, fire, and emergency medical units. The emergency services research can be divided into six areas: (1) relationship between measures of effectiveness and system performance; (2) number of units needed in a region; (3) deterministic location models; (4) probabilistic location, districting and dispatching models; (5) staffing and scheduling; and (6) emergency service mergers. The chapter describes the characteristics of fire, police, and emergency medical problems. The pitfalls of traditional approaches to these problems are discussed. Deterministic location models assume that the nearest unit to a call for service is always available. In contrast to determining the number of units needed in a region, selecting particular locations or districts is more difficult since the responsibilities and behavior of individual units must be considered. The chapter also describes the models and the direction of future work.

Using a Fractional Factorial Design to Increase Direct Mail Response at Mother Jones Magazine

Experimental design methods have been widely applied to problems in manufacturing, but little attention has been given to applying these tools to service problems. In the world of direct marketing, the traditional approach is called A/B testing and involves changing one factor at a time. In this article we address the problem of improving direct mail response at Mother Jones magazine, employing a 16-run two-level fractional factorial design that tests seven factors simultaneously. We compare this design to other possible design choices, discuss sample size determination, and show how we labeled factors to isolate likely two-factor interactions. We discuss the results and conclusions of our study and the benefits to Mother Jones magazine.

A Plackett-Burman Experiment to Increase Supermarket Sales of a National Magazine

This paper describes and discusses a Plackett-Burman experiment aimed at increasing supermarket sales of a top-selling national magazine. The experiment involved 10 factors relating to in-store advertising and the location of the magazine within the store. We discuss issues including choice of factors, alternative designs, sample-size determination (number of test stores and the length of the test period), and the analysis of the resulting data. We show the large benefits that accrued from our approach of testing many factors simultaneously compared to the common industry practice of sequentially testing one factor at a time. We also discuss the potential opportunities of applying this approach to other service problems.

Quality Activities and Quality Performance of Japanese and U.S. Firms in the Printed Circuit Board Industry

In the last 10 to 15 years, U.S. manufacturing firms spurred by their Japanese competitors have taken a renewed interest in quality management. Under the banner of statistical process control (SPC) or total quality management (TQM), there has been an emphasis on statistical methods, worker involvement, and top management support. This wave of interest in quality gained momentum in the mid-1980’s following a similar spurt of attention to cycle time reduction through Just-In-Time (JIT) manufacturing.

Assessing hearing conservation program effectiveness: Results of a multisite assessment

Objectives: Noise-induced hearing loss (NIHL) remains one of the most prevalent occupational diseases. Occupational hearing conservation programs (HCPs) can reduce the risk of NIHL, but there remains no consensus on assessing HCP effectiveness. We conducted a multisite, mixed-method assessment of HCP programs. Methods: At 13 manufacturing plants, we performed assessments, including interviews with program staff and worker focus groups and surveys. We analyzed the association between these assessments and age-corrected NIHL rates. Results: Only a few items from the HCP staff interviews correlated with NIHL rates. For the employee survey, management commitment to NIHL prevention and being counseled about NIHL were strongly associated with NIHL rates. Conclusion: Management commitment and counseling of workers about NIHL may be key factors in program effectiveness. A combination of qualitative and quantitative methods appears to be useful for assessing HCPs.