John F. Kros

Data mining and the impact of missing data

The actual data mining process deals significantly with prediction, estimation, classification, pattern recognition and the development of association rules. Therefore, the significance of the analysis depends heavily on the accuracy of the database and on the chosen sample data to be used for model training and testing. Data mining is based upon searching the concatenation of multiple databases that usually contain some amount of missing data along with a variable percentage of inaccurate data, pollution, outliers and noise. The issue of missing data must be addressed since ignoring this problem can introduce bias into the models being evaluated and lead to inaccurate data mining conclusions. The objective of this research is to address the impact of missing data on the data mining process.

Impact of just‐in‐time inventory systems on OEM suppliers

Purpose – To analyze the impact of the adoption of just‐in‐time (JIT) production systems by different equipment manufacturers (OEMs) on the inventory profiles of their suppliers.Design/methodology/approach – The research is designed to examine five financial measures of inventory management performance over the years 1994‐2004. Three specific industry sectors where OEMs have adopted and implemented JIT principles are studied. These sectors include the automotive, electronics, and aircraft industries. A one factor analysis of variance is employed to the five hypotheses and Tukeys post‐hoc test is used to interpret statistical pairwise differences between level means.Findings – Overall, the research finds that OEM suppliers in the automotive, electronics, and aircraft sectors have shown mixed results in the impact JIT implementation has had on inventory performance measures.Research limitations/implications – The research focuses on three industrial sectors over approximately a ten year time frame that may...

Economic design of Xbar control charts with continuously variable sampling intervals

In the context of economic design, a two-stage model is proposed that utilizes continuously variable sampling intervals. Specifically, each successive sampling interval is determined by the extremity of the latest sample. Modeling the situation as a Markov chain, the hourly cost is developed for any arbitrary set of design parameters. This proposed approach is found to be more economically desirable, possessing a smaller average out-of-control production time, when compared with a standard two-stage approach where the sampling interval alternates between two fixed values. Copyright

Overbooking Increases Patient Access at East Carolina University's Student Health Services Clinic

The health-care clinic presented in this study experienced significant numbers of patients who failed to arrive for their scheduled appointments (no-shows). The cost of reducing patient access at this clinic because of no-shows is estimated to exceed

Technology emergence between mandate and acceptance: an exploratory examination of RFID

400,000 annually. An interdisciplinary quality-improvement team developed a novel health-care overbooking model that includes the effects of employee burnout. This model estimates the nonlinear nature of the costs associated with medical-provider burnout caused by overbooked appointments that exceed clinic capacity. Several key East Carolina University clinical staff members had been skeptical about the value of overbooking. The model was instrumental in convincing them that implementing an overbooking process would benefit patients and the organization. The clinic, which subsequently implemented such a process, attributes a savings of

Effects of the neural network s-sigmoid function on KDD in the presence of imprecise data

95,000 per semester to the initiative.

An assessment of supply chain managers' trust in online auctions

Purpose – The purpose of this paper is to focus on radio frequency identification (RFID) acceptance and examine three understudied drivers: a companys satisfaction with existing logistics technologies, its logistics technology readiness (technology optimism and technology innovativeness), and relationship hostage position. The proposed conceptual model also investigates the impacts of RFID acceptance and these three antecedents on the companys logistics performance.Design/methodology/approach – Survey data were collected from the members of three professional associations in the USA. Structural equation modeling was used to test the proposed relationships.Findings – According to the analysis results, a companys satisfaction with existing technology has negative impact on RFID acceptance, and technology readiness has positive impact on RFID acceptance. However, the relationship between a companys hostage position and RFID acceptance was found to be only partially significant. Also, the positive relatio...

Teaching communication in an MBA operations research/management science course

This research explores a specific step in the Knowledge Discovery of Databases (KDD) process, Data Mining. The actual data mining process deals significantly with prediction, estimation, classification, pattern recognition and the development of association rules. Therefore, this analysis depends heavily on the accuracy of the database and on the chosen sample data to be used for model training and testing. Data mining is based upon searching the concatenation of multiple databases that usually contain some amount of missing data along with a variable percentage of inaccurate data, pollution, outliers and noise. The issue of missing data must be addressed as ignoring this problem can introduce bias into the models being evaluated and lead to inaccurate data mining conclusions. The objective of this research is to address the Effects of the Neural Network s-Sigmoid Function on KDD in the Presence of Imprecise Data using a three factor ANOVA test and Tukeys Honestly Significant Difference statistics.This research investigates the accuracy and impact of Data Imputation Methodologies that are employed when a specific Data Mining algorithm is utilized within a Knowledge Discovery In Databases (KDD) process. This study will employ certain Knowledge Discovery processes that are widely accepted in both the academic and commercial worlds. This work includes testing the impact of missing data on the Neural Network s-Sigmoid Transfer Function type in the Data Mining process, by experimenting with three factors: imputation method, data set size, and level of data missingness.

An exploration of quality management practices, perceptions and program maturity in the supply chain

Purpose – The purpose of this study is to empirically test the levels of trust supply chain managers exhibit regarding online auctions. The study seeks to test the hypotheses that supply chain managers with higher levels of trust in online auctions would exhibit more familiarity with online auctions, more positive relationships with suppliers, stronger focus on cost management, a stronger focus on revenue enhancement, and greater levels of participation.Design/methodology/approach – A review of the extant literature provided the basis for the development of an online pilot study that was e‐mailed to 100 supply chain managers. The results of the pilot study were used to develop an online survey instrument that was e‐mailed to 2,313 current members of the Council of Supply Chain Management Professionals (CSCMP) and the Institute of Supply Management. A total of 213 usable questionnaires were returned, resulting in a 9.2 percent response rate.Findings – The data were analyzed via t‐test, MANOVA and ANOVA. Th...

A decision support system for quantitative measurement of operational efficiency in a blood collection facility

This paper discusses the inclusion of a writing requirement within a first-year Operations Research/Management Science (OR) course of a Masters of Business Administration (MBA) curriculum. The inclusion of a writing requirement addresses the importance of communication in OR. It is the authors’ supposition that many of the recent criticisms of OR teaching have their roots in poor OR communication. Although OR faculty members can well argue relative values, we have failed to adequately prepare our students to communicate relevant insights from OR studies. The integration of a writing requirement, consisting of short executive summary-style reports and a major term project, within an MBA OR course can do much in promoting further OR study and OR work. This paper describes our experience in instituting a writing requirement within a first-year MBA course in OR.