Using Shewhart and CUSUM Charts for Diagnosis with Count Data in a Vendor Certification Study

Abstract

Problem: We present a case study concerned with the problem of reducing errors by vendors who process credit card applications for a large credit card bank. More specifically, we show the results of a process performance evaluation and process capability analysis that was used to help identify and solve the problem of excess variability. These tasks are sometimes called Phase I studies. We conclude with a discussion of implications for process monitoring and subsequent process improvements, sometimes called Phase II studies.Approach: Quality of the credit card processing task was measured with count data and the counts were low; this is typical of many service operations today. Our approach was to use sequence plots and CUSUM sequence plots of the data and Shewhart np and CUSUM control charts to explore the count data drawn from the four vendors that process credit card applications for this bank. We found that using both Shewhart and CUSUM methods helped us to better accomplish the Phase I study.Results: The results of our analysis indicated differences in quality between vendors and within vendors over time. Subsequent analysis showed that the differences were related directly to training and workers' experience. Two interventions were made subsequent to the process capability analysis; these were improved training and interventions to reduce turnover.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 13 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.