Knowledge that Transforms

We describe a set of models that are used to manage the product and process design in a mass placement printed circuit board (PCB) assembly cell. Our models can be divided into two categories. First, we characterize the cell design problem and develop models to design an efficient assembly cell. Second, we present models for optimizing the different operational aspects of the assembly cell. These models were developed to assist the managers of a large electronic manufacturing firm in establishing mass placement PCB assembly cells.

We propose an agglomerative heuristic cluster analysis framework for application to the part‐family and machine‐cell formation problems associated with group technology. This framework addresses the notion of concurrently forming clusters of parts (families) and machines (cells) based on natural between‐part and between‐machine relationships and the strength of association relating pairs of parts with pairs of machines. An illustrative model is presented and operational aspects demonstrated using a small problem.

Underlying standard control‐chart methodology is the assumption of independent and identically distributed (IID) random variables. However, autocorrelation and other time‐series effects frequently occur in manufacturing and service quality applications. As a result, it has been suggested that standard methods be extended by using time‐series modeling. A little statistical analysis can go a long way, especially when time‐series effects are strong enough to frustrate efforts directed toward discovery of special causes. But a practical limitation on the use of time‐series modeling is that its implementation requires some sophisticated statistical skills, whereas standard control charts require only elementary statistical knowledge. The choice raises the questions of management philosophy, statistical techniques, and computation.

Redesigning and improving business processes to better serve customer needs has become a priority in service industries as they scramble to become more competitive. We describe an approach to process improvement that is being developed collaboratively by applied researchers at US WEST, a major telecommunications company, and the University of Colorado. Motivated by the need to streamline and to add more quantitative power to traditional quality improvement processes, the new approach uses an artificial intelligence (AI) statistical tree growing method using customer survey data to identify operations areas where improvements are expected to affect customers most. This AI/statistical method also identifies realistic quantitative targets for improvement and suggests specific strategies predicted to have high impart. This research, funded in part by the Colorado Advanced Software Institute (CASI) to stimulate profitable innovations, has resulted in a practical methodology used successfully at US WEST to help set process improvement priorities and guide resource allocation decisions throughout the company.

Computer‐integrated manufacturing implementation is often hindered by human resource issues like stress. By focusing on one type of potential human obstacle to the integration of islands of automation in our exploratory case study, we examine the sources of stress and the lack of job control experienced in three functional departments: computer‐aided design and manufacturing, manufacturing planning and control, and computer numerical control/robot manufacturing; we also suggest management interventions to alleviate stress in each group of workers. We used three methods to collect data: self‐reporting, observation at the job level, and physiological screening. Our results indicated little difference in stress levels among the three groups. However, we found that each group faced a different set of pressures and exercised different levels of control over their jobs. We suggest specific managerial action to overcome the most critical pressures.

We give a tutorial on bottleneck dynamics. Bottleneck dynamics is a scheduling framework that uses approximate dual resource prices to make decentralized decisions. The basic idea is to establish a price for a resource as a function of the set of jobs that need to be processed by the resource. Tasks are then sequenced according to a cost/benefit ratio. Starting with one resource sequencing problems, we describe how priorities for jobs can be developed and how they translate into resource prices. We then describe how resource prices can be approximated in a multiresource situation and how lead times which are critical for these approximations can be accurately computed. We also describe a number of studies that have shown bottleneck dynamics to be an effective approach in several different problem areas.