Reducing Not-In-Good-Order Rates in Insurance and Structured Settlement Applications: A Simulated Lean Six Sigma Intervention with Statistical Process Control and Modelled Cycle-Time and Rework Outcomes

Authors

  • Krishna Deep Valluru

Keywords:

not-in-good-order, Lean Six Sigma, DMAIC, statistical process control, p-chart, individuals chart, attribute measurement system analysis, operational definition, insurance operations, synthetic data, simulation study, method demonstration.

Abstract

The title's reference to measured outcomes means measured on the synthetic population described in Section III. With that stated first, the substance is this. A not-in-good-order application is one that cannot proceed through normal processing because required information, signatures, documents, forms or eligibility details are missing, incorrect or inconsistent, and must leave the standard workflow for exception handling. The operational literature on such defects in insurance and structured settlement intake is, so far as a systematic search can establish, empty: the entire evidence base is vendor and trade material, with no refereed treatment at all. This paper specifies and exercises an analysis protocol for that problem end to end, on a synthetic population of 3,700 applications across twelve monthly subgroups. The protocol has a deliberate ordering. The operational definition is repaired first, because inconsistent reviewer classification of borderline defects is a measurement system defect rather than a process defect, and a baseline measured through an unreliable classifier cannot anchor anything that follows. Only then are process changes introduced, in four stages: capture redesign, validation at point of entry, standardised training and decision rules, and upstream automation of the controls that proved stable. Monitoring pairs a p-chart with control limits recalculated for each monthly subgroup, since a monthly census varies between roughly 280 and 330 applications, with an individuals and moving range chart on monthly mean cycle time. The paper's substantive argument concerns a root cause that could not be eliminated. Suitability and replacement exceptions depend on regulatory requirements, product rules and approvals outside the operating team's control, and their incidence does not move across the two periods. The response is to stop attacking the defect and attack its downstream cost instead, which is a distinction the improvement literature, framed almost uniformly around elimination, does not treat. What this paper can claim is that the protocol detects a shift of known magnitude under stated conditions and is internally coherent. What it cannot claim, and does not, is any effect in any real operation.

Downloads

Published

2026-09-28

How to Cite

Valluru, K. D. (2026). Reducing Not-In-Good-Order Rates in Insurance and Structured Settlement Applications: A Simulated Lean Six Sigma Intervention with Statistical Process Control and Modelled Cycle-Time and Rework Outcomes. International Journal of Artificial Intelligence and Machine Learning, 6(12s), 1723–1733. Retrieved from https://www.svedbergopen.com/index.php/ijaiml/article/view/3181