John R. English

Process capability analysis—a robustness study

Abstract The robustness of two popular process capability ratios, Cp and Cpk, when the random process being observed departs from normality is analysed. The distributions of estimated process capability ratios are derived and used as a basis for validation of large-scale simulation studies in an examination of departures from normality. The simulation studies and analytical results provide a basis for recommended procedures. It is recommended that the impact of process distributions be considered before using popular process capability indexes, due to the lack of robustness when departures from normality are ecountered. As an extension of our findings, we consider the Taguchi loss function as a generalization to process capability analysis regardless of the underlying population distribution.

Designing New Products

In many industries, the success or failure of a new product is largely determined in the design phase of the project. Consequently, manufacturers should strive to develop integrated system design tools to help ensure that new products are readily producible according to manufacturing or assembly capability. Introduces a systems integration strategy called “design to fit an existing environment” (DFEE). The purpose of this DFX sub‐strategy is to integrate capacity and product mix issues into design strategies so that it becomes possible to design new products which are more globally optimal from a systems standpoint. The result of using DFEE is that better product time‐to‐market targets can be established, and products can be designed to take advantage of existing slack manufacturing capacity, thus balancing assembly flow and reducing capital expenditures.

Detecting changes in autoregressive processes with X¯and EWMA charts

The traditional use of control charts necessarily assumes the independence of data. It is now recognized that many processes are autocorrelated thus violating the fundamental assumption of independence. As a result, there is a need for a broader approach to SPC when data are time-dependent or autocorrelated. This paper utilizes control charts with fixed control limits for residuals to monitor the performance of a process yielding time-dependent data subject to shifts in the mean and the autocorrelation structure. The effectiveness of the framework is evaluated by an average run length study of both X¯ and EWMA charts using analytical and simulation techniques. Average run lengths are tabulated for various process disturbance scenarios, and recommendations for the most effective monitoring tool are made. The findings of this research present motivation to extend the traditional paradigms of a shifted process (e.g., mean and/or variance). The results show that decreases in the underlying time series parameters are practically impossible to detect with standard control charts. Furthermore, the practitioner is motivated to employ runs rules since the runs are more likely with time-dependent observations.

Forecasting freight demand using economic indices

This paper describes the results of an effort to predict future freight volume in the truckload (TL) trucking industry. The approach involves the use of stepwise multiple linear regression models that relate freight volume to a variety of economic indicators. The models are built using a large set of actual freight data provided by J.B. Hunt Transport (JBHT), one of the world’s largest TL carriers. The data was first analyzed using the overall set of national data, and then for specific industrial and regional segments. The overall results of these analyses should prove useful to a wide variety of transportation and logistics operations.

Economic designs of constrained EWMA and combined EWMA-X¯ control schemes

Abstract This research presents a comparison between the cost performance of the Exponentially Weighted Moving Average EWMA) and the combined EWMA-X¯ control chart schemes. In particular, we explore the impact of constraining the in-control average run length on the optimal cost performance of both schemes. Methodologically, we incorporate traditional expected cost models and study the robustness of the two approaches. In general, there appears to be minimal motivation to combine the use of both charts within the same application. The cost model for the combined chart is not a well-behaved function, and yields varying optimal parameters when the in-control average run length is constrained.

A discretizing approach for stress/strength analysis

This paper implements and evaluates a discretizing approach for estimating the reliability of systems for which complex functions define strength or stress and where the derivation of reliability exceed analytic techniques. The discretizing approach predicts system reliability with reasonably high accuracy. Specifically, there is little difference in the accuracy of predictions for three engineering problems when compared to simulation results. The reliability predictions are near the 95% confidence intervals of the simulation results and are best in the high reliability and low reliability regions. The small errors observed are attributed to the estimation errors of the discretizing approach. The mid-range reliability values (e.g. 50% reliability) are not generally of interest in engineering applications, and even for these value, the errors are small. There is little improvement in increasing the number of points in the pmf from 3 to 6. Due to this small difference, 3 discretizing points are recommended for reliability predictions when computational ease is of concern and limited to 4 points when more accurate reliability predictions are required. This paper models three systems and evaluates the robustness (departures from assumed distributions) of the discretizing approach. The discretizing approach is not too sensitive to departures from the assumed distribution of the underlying random variables regions are accurately estimated.

Modeling and analysis of EWMA control schemes with variance-adjusted control limits

A stochastic model of the EWMA control chart employing the variance-adjusted control limits is presented. The average run length for various shifts in the process mean is approximated by using a Markov chain approach. Following this, an alternative basis for the design of EWMA schemes is proposed. Computational results are presented and the impact of employing the variance-adjusted control limits on the average run length is investigated. Finally, conditions that permit the use of the asymptotic control limits in place of the variance-adjusted control limits are also identified.

Quality monitoring of continuous flow processes

Abstract In this research, existing quality control techniques for monitoring continuous flow processes are evaluated. Autocorrelation is a common characteristic of continuous flow process data, and the effect of the autocorrelated data is modelled as an autoregressive time series model of order one or two. The process is simulated on the computer for various process parameters, and the effectiveness of a given statistical process control technique for detecting known process disturbances is evaluated by determining the average run length. Due to the limitations of existing statistical process control techniques, a recursive Kalman filter is proposed as an alternative for eliminating the autocorrelation from the process data. The modelled manufacturing process, the computer simulation results, and the recursive Kalman filter are summarized in this paper.

System reliability modeling considering the dependence of component environmental influences

A system reliability modeling procedure is described and demonstrated to accommodate the case when component failure times are statistically correlated because of the shared environmental exposure of components within a system. When component failure times are correlated, independence assumptions are not valid, and thus, many common reliability modeling practices are inappropriate. If component reliability is influenced by environmental exposure, then the components within a system are likely to have correlated time-to-failures because all components within a system are influenced similarly by the system-level environmental stress profile. This scenario is often overlooked when failure data is analyzed for a homogeneous population of parts that have experienced nonhomogeneous usage profiles. The model presented here is based on proportional hazards models for component reliability and discretized approximation of the joint probability density function for system environmental stress variables. The discretization approach is mathematically convenient, accurate and offers several pragmatic advantages over alternative computation approaches. A hypothetical three-component series system is analyzed, and the results are compared to two common approximations: (1) component independence assumption and (2) use of environmental stress average values. The results indicate that the described approach is convenient and has the potential to be scaled-up to large problems.

A modular simulation approach for automated material handling systems

Abstract In this paper, the authors present a modular simulation approach for the evaluation of automated material handling systems (AMHSs). The approach involves the use of high level modularity so that alternative AMHS scenarios can be evaluated using identical or nearly identical source code. The approach is implemented using the AutoMod simulation platform, and the efficacy of the strategy is demonstrated through a case study with the United States Postal Service. Results indicate that the approach is viable and that it is of value for the evaluation of AMHS scenarios in general.