Using frailty models to account for heterogeneity in multistage manufacturing and service processes

Statistical process control has been widely applied to manufacturing and service operations with the aim of monitoring and improving the reliability of products. The existing monitoring procedures were introduced following the assumption that a single-stage process with independent quality characteristic is under consideration. However, in multistage processes with dependent variables, quality characteristics of interest should be optimally monitored only after they have been adjusted for the effect of influential covariates. In general, it is impossible to include all relevant covariates because measuring such values entails great financial costs. The neglect of such covariates results in having unobserved heterogeneity which dampens the detection ability of the monitoring procedure. The more complicated picture arises when the values corresponding to the reliability-related quality variable are censored due to the time and cost constraints. Thus, to deal with the effect of observed and unobserved covariates together with the censoring issue, the frailty and the proportional hazard models are used and some model-based monitoring schemes are devised. The surveillance procedures are proposed in both the presence and absence of a censoring mechanism. The performance analysis shows that the monitoring procedure based on the cumulative sum chart is superior in detecting shifts while the exponentially weighted moving average chart is effective in some cases.     

Influence factors for operational control and compensation in professional service firms

Professional service firms have distinct operational challenges due to the type of work that is transacted by the employees of these firms, and due to the nature of the employees themselves. In this paper, we develop and present factors that influence professional service operations in firms and compensation structures for professional service providers. We establish professional service influence factors, which we posit will impact agency relationships in professional service firms. That is, we hypothesize that professional service influence factors (PSIFs) will moderate the effect of task programmability and outcome measurability in predicting the use of behavior- or outcome-based compensation schemes (control strategy). Logistic regression is used on data provided by 192 professional service providers in order to examine the impact of the agency variables and moderating factors on control strategies. The results indicate that company- and profession-based factors have moderating effects on task programmability when predicting control strategy. We discuss the implications of our findings.     

Operations related groups (ORGs): A clustering procedure for production/inventory systems

Many production/inventory systems contain thousands of stock keeping units (SKUs). In general, it is not computationally (or conceptually) feasible to consider every one of these items individually in the development of control polices and strategies. Our objective here is to develop a methodology for defining groups to support strategic planning for the operations function. Accordingly, such groups should take into consideration all product characteristics which have a significant impact on the particular operations management problem of interest. These characteristics can include many of the attributes which are used in other functional groupings and will most certainly go beyond the cost and volume attributes used in ABC analysis.The ORG methodology is based on statistical clustering and can utilize a full range of operationally significant item attributes. It considers both statistical measures of discrimination and the operational consequences associated with implementing policies derived on the basis of group membership. The main departure of this analysis from earlier work is: 1) the approach can handle any combination of item attribute information that is important for strategy purposes, 2) management's interest in defining groups on the basis of operational factors can be accommodated, 3) statistical discrimination is considered directly, 4) group definition reflects the performance of management policies which are based (in part) on group membership, and 5) the method can be applied successfully to systems with a large number of SKUs.The specific application which motivated development of the ORG methodology was an analysis of distribution strategy for the service parts division of a major automobile manufacturer. The manufacturer was interested in developing optimal inventory stocking policies, which took into account the complexities of its multiechelon distribution network, supplier relationships and customer service targets for each market segment. This manufacturer stocked over 300,000 part numbers in an extensive network with approximately 50 distribution centers and thousands of dealer locations (i.e., 1.5 million SKU/ location combinations). The results of this application indicated that the advantage of using operationally relevant data for grouping and for defining generic, group-based policies for controlling inventory can be substantial. The ORG methodology can be of value to operations managers in industries with a large number of diverse items.     

No excuses: forecasting can be managed

Originally written for manufacturers, the forecasting concepts presented in this article are easily applied to a hospital or health care facility with a supply function or central distribution center. Forecasting can offer a method of managing the control of medical supplies, pharmaceuticals, or maintenance items. A statistical based forecasting approach with seasonality testing could be useful in analyzing patient census data, calculating trends, and then recommending staffing levels. Cost containment and efficiency of operations are priorities in any industry. Forecasting the demand of medical supplies and services in today's hectic health care environment can have a significant effect on both customer service and financial results. The bottom line is happy patients and healthy profits.     

Combining density forecasts using focused scoring rules

We investigate the added value of combining density forecasts focused on a specific region of support. We develop forecast combination schemes that assign weights to individual predictive densities based on the censored likelihood scoring rule and the continuous ranked probability scoring rule (CRPS) and compare these to weighting schemes based on the log score and the equally weighted scheme. We apply this approach in the context of measuring downside risk in equity markets using recently developed volatility models, including HEAVY, realized GARCH and GAS models, applied to daily returns on the S&P 500, DJIA, FTSE and Nikkei indexes from 2000 until 2013. The results show that combined density forecasts based on optimizing the censored likelihood scoring rule significantly outperform pooling based on equal weights, optimizing the CRPS or log scoring rule. In addition, 99% Value‐at‐Risk estimates improve when weights are based on the censored likelihood scoring rule.     

The quality movement in hospital care

The purpose is to introduce the demand for the quality movement practice in hospital care. We show both the need and application of quality monitoring, especially the need monitoring activities having auto correlated data flows of which there are many in the hospital environment. The goal is to control the flow of quality care data in the dynamic behavior of these systems of acre in hospitals. These monitoring systems are designed to control and improve changes in the hospital care environment.     

Strategic operations management decisions and professional performance in U.S. HMOs

Professional service organizations are becoming an increasingly important segment of the service sector in the U.S. but little attention has been paid to the management of these organizations, particularly in relation to technical performance. This paper reports the findings of a survey of managers of Health Maintenance Organizations (HMOs) which related operations management decisions about organization, workforce management, quality management and planning and control, to clinical process and outcome performance.This research demonstrates that managerial decisions do affect clinical performance in HMOs. These findings have implications for both researchers and practitioners. For researchers, the study provides a framework for future research on the important link between management decisions and technical performance in professional service organizations. For practitioners, this study suggests that technical performance will be enhanced when professional work is appropriately managed.     

Comparisons of control schemes for monitoring the means of processes subject to drifts

Although statistical process control (SPC) techniques have been focused mostly on detecting step (constant) mean shift, drift which is a time-varying change frequently occurs in industrial applications. In this research, for monitoring drift change, the following five control schemes are compared: the exponentially weighted moving average (EWMA) chart and the cumulative sum (CUSUM) charts which are recommended detecting drift change in the literature; the generalized EWMA (GEWMA) chart proposed by Han and Tsung (2004) and two generalized likelihood ratio based schemes, GLR-S and GLR-L charts which are respectively under the assumption of step and linear trend shifts. Both the asymptotic estimation and the numerical simulation of the average run length (ARL) are presented. We show that when the in-control (IC) ARL is large (goes to infinity), the GLR-L chart has the best overall performance among the considered charts in detecting linear trend shift. From the viewpoint of practical IC ARL, based on the simulation results, we show that besides the GLR-L chart, the GEWMA chart offers a good balanced protection against drifts of different size. Some computational issues are also addressed.     

Asymmetry and nonstationarity for a seasonal time series model

Tests for symmetry and seasonal unit roots are developed for an extended model of Hylleberg et al. (1990. Seasonal integration and cointegration. Journal Econometrics 44, 215–238.) which can represent both partial seasonal unit roots and threshold effects. Methods based on ordinary least squares (OLS) estimation and instrumental variable (IV) estimation are proposed and compared. For adjusting mean functions, ordinary mean adjustment and recursive mean adjustment are both considered. Several tests are constructed from various combination of estimation schemes and mean adjustment schemes. Among the tests, the tests based on IV-estimation are recommended because they have very simple limiting null distributions and have finite sample power properties comparable to those based on the OLSE. The recommended tests are applied to a US unemployment rate data set and find evidences for both nonstationarities associated with zero frequency and threshold effects.     

Investigation of causality based on complex use of statistical methods (case study of social research)

The article deals with the problem of revealing cause–effect relations in the social research by a complex of statistical methods. The ways for solving the problem for experimental and non-experimental data are regarded. In the research the role of the statistical methods in formation of causal relation hypotheses is revealed and the methods of hypotheses verification in conditions of experiment or upon latent structure modelling are discussed. The possibilities of the use of the factor analysis and the analysis of variance in revealing cause–effect relations are also considered. Special attention is paid to the interpretation of the analysis results depending on the character of the empirical data and the conditions they were obtained in. The approaches to causal relations revealing based on the development of the schemes of the complex analysis of the empirical data by the statistical methods aimed at investigation of the cause and effect relation in a long-term research are suggested. We take into account the hierarchical structure of the schemes reflecting the structure of the subject of cognition, the specificity of which requires the use of “rigid” and “flexible” models in combination. Its variant could be the method of structural equation modelling. The article gives an example of the research of causal relations in the assessment of the quality of education services in higher institution. Each stage of the assessment is provided with building of certain “rigid” or “flexible” mathematical models. Their combination allows using the obtained quality parameters as an instrument of regulation of interaction between education service providers and clients.     

项目分包质量管理的策略与方法

项目分包是一种重要的生产合作方式,已为越来越多的企业所采用。研究项目分包质量管理的策略与方法,具有重要的理论和实践意义。本文首先根据项目分包的特点与分包过程,提出了项目分包质量管理的三阶段策略;其次,应用分包管理理论、项目质量管理方法,研究了项目分包各阶段质量管理的方法。     

QMS导入世界特殊奥运会及其实践初探

世界特殊奥运会将于2007年首次在中国、在上海举办,为了保证本届特殊奥运会的成功举办,需要建立一套特殊奥运会质量管理标准,对特殊奥运会服务过程各个关键环节实施质量管理和控制。本文从特殊奥运会流程分析入手,确定特殊奥运会服务管理内容和关键服务流程,然后应用相关矩阵法对关键流程进行二级展开和重要度排序,据此识别特殊奥运会服务关键控制点,并总结出关键控制点识别的6项原则。针对服务关键控制点所体现的质量特性,提出相应管理措施。     

Data center supply chain configuration design: A two-stage decision approach

Data centers are special-purpose facilities that enable customers to perform cloud based real-time online transactions and rigorous computing operations. Service levels of data center facilities are characterized by response time between query and action, which to a large extent depends on data center location and data travel distance. Another aspect of service level is resource up-time availability, which is determined by data center configuration. Data center location and configuration decisions are, therefore, of great significance to ensure uninterrupted operations in customers of manufacturing and service industries relying on cloud-based computing resources. In this study, following a grid-based location approach, we present two mixed integer linear programming models for capacitated single-source data center location-allocation problems. The first model provides optimal locations, capacities and configurations of data centers, and allocation of demands to open facilities when there is no existing facilities in the region. Our second model considers the decision problem of meeting new demand when the existing demand is met by the already opened facilities. We term these newly arrived demand as replication demand, which results either from emergence of new users of existing customers at distant locations in the future, or as a means of increasing data resilience by creating data replication as a backup. To solve the decision problem for meeting primary and replication demand optimally, we propose a two-stage decision algorithm. The algorithm provides optimal locations, capacities and configurations for new data centers, capacity addition decisions to the existing facilities and subsequent allocation of demands. Both models and solution algorithm are implemented using AMPL programming language and solved with CPLEX solver. The models are found to be scalable and capable to provide high quality solutions in reasonable time.     

The non-uniform pricing effect of employee stock options using quantile regression

Issuing employee stock options (ESOs) transfers equity claims from current stockholders to employees, and thereby dilutes existing shareholder interests. Because employees are motivated to exert additional effort toward better performance, the value of transferred ownership claims proxied by ESO expense represents a cost of generating firm value. There are several econometric issues, most notably including the fact that the disclosed ESO expense is an endogenous variable. Without controlling for the simultaneity problem, inferences based on results from OLS analyses may be misleading. More importantly, a considerable amount of ESO expense data is censored at zero. Such a censoring problem can make the population distribution severely skewed, resulting in estimation bias. Therefore, we need to take into account the censored data issue. No prior studies have considered these two issues simultaneously. Failure to control for both censoring problem and endogeneity could explain the inconsistent results documented in prior studies. In this paper, we use the two-stage quantile regression (QR) proposed by Amemiya (1982) and Powell (1983) to examine possible nonlinear relationships, especially whether conditionally higher-stock price (or better performing) firms show a stronger negative pricing effect of ESO expense (that is, the relation between ESO expense and share price) than conditionally lower-share price firms. Our results suggest that the linear regression model greatly underestimates this negative pricing effect at higher quantiles, so the nonlinear relationship is obscure when using the standard linear model. We also consider alternative interpretations as to why heterogeneity exists in the pricing effect of ESO expense and assess whether our results concur with these explanations.     

Hotelling’s <Emphasis Type="Italic">T</Emphasis><Superscript>2</Superscript> control chart with two adaptive sample sizes

Some quality control schemes have been developed when several related quality characteristics are to be monitored. The familiar multivariate process monitoring and control procedure is the Hotelling’s T ² control chart for monitoring the mean vector of the process. It is a direct analog of the univariate shewhart [`(x)]{\bar{x}} chart. As in the case of univariate, the ARL improvements are very important particularly for small process shifts. In this paper, we study the T ² control chart with two-state adaptive sample size, when the shift in the process mean does not occur at the beginning but at some random time in the future. Further, the occurrence time of the shift is assumed to be exponentially distributed random variable.     

Design of financial incentives and payment schemes in healthcare systems: A review

In this paper, we review the existing studies on the design of financial incentives and payment mechanisms for healthcare service providers, which have been published in the literature of operations research & management science. We classify these studies in two ways: first, by the care provider focused on: physicians or hospitals; and next, by categorizing payment/incentive schemes as retrospective or prospective models under each care provider group. The problem setting, including the geographic background and research methodology, are briefly discussed for each paper, as are the findings with regard to accessibility, quality, efficiency, and the facilitation of integration and collaboration, if reported. We also provide a detailed discussion of potential research problems that would offer promising avenues for future study.     

Non-parametric maximum likelihood estimation of censored regression models

This paper presents a consistent estimator of a censored linear regression model which does not require knowledge of the distribution of the error term. The estimator considered here applies Duncan's (1982) suggestion that the likelihood function for the censored regression model be treated as a functional of both the unknown regression vector and the unknown error distribution. Our estimator is the majorizing regression vector for this non-parametric likelihood functional. We find conditions which ensure the consistency of the NPMLE. The paper concludes with the results of Monte Carlo experiments which show the NPMLE to be more efficient than Powell's Least Absolute Deviations (LAD) estimator, particularly when the fraction of censored observations is large and the sample size is small.     

A pull system for delegating knowledge work

A pull system using kanban is commonly used in manufacturing settings to efficiently control the flow of goods. Its success in service operations is limited to processes similar to production lines when the output is repetitive. This paper examined how well a pull system for delegating non-repetitive output performed in an experimental setting of knowledge work, similar to what is found in many services. Results indicated that performance, as measured by completion time for cognitive tasks, improved under a pull as opposed to a push system of delegation. The improvement occurred with no change in output quality, stress levels, or satisfaction. Portions of this data were presented at the Academy of Management Conference, Philadelphia, PA, August, 2007.     

Customer-centered measurement of service operations: a B2B case study

This study examines how the customer can be utilized in performance measurement and evaluation of complex engineered service operations in the business-to-business environment. The framework put forth by Ukko and Pekkola for a customer-centered approach to performance measurement in service operations is used, focusing on the special nature of service processes and the customer’s perspective. The results of this case study show that in non-interactive customer-related processes, performance evaluation can be a viable alternative to performance measurement. In addition, enabling the customer to participate in the evaluation and measurement of service operations will generate not only new measures but also new services.     

CAST污水处理工艺自动控制系统设计

我国城市污水处理正在实现跨越式发展。由于其环境条件的特殊性和监控项目太多等因素,要采用人工现场操作难度极大。为保证其运行稳定高效,必须采用污水处理自动化技术。本文结合我们设计改造并成功的案例,介绍了CAST污水处理工艺控制系统,对各工序控制要点及控制时序进行了描述,并就系统设计和运行过程中有关问题进行了分析。