Root cause analysis of product service failures in design-A closed-loop lifecycle modelling approach

A number of industries including aerospace, telecom and automotive incur warranty and product return costs due to product malfunctions in service, which can also negatively impact customer satisfaction and loyalty. Product failures, which occur in service, are often caused by root causes at design or at manufacturing phases. Therefore a novel inter-loop modelling framework is needed which takes information from different phases to determine root causes and corrective actions. This goes beyond current intra-loop methods such as design optimization, statistical process control etc., which uses information from a single phase to address failures in the same phase. However, inter-loop modelling poses the challenge of integrating heterogeneous data from different phases of lifecycle with product and process models to determine failure root causes and corrective actions. To deal with failures in service, this paper proposes Closed-loop Lifecycle Modelling approach, specifically integrating information from service and design. Related work on fault diagnosis and corrections is also reviewed in the context of intra and inter loops of product lifecycle. The proposed methodology addresses root cause analysis (RCA) of service failures caused due to dimensional variations of product features. RCA identifies critical geometric features of internal components, which affects dimensional variations of product features. This is done by integrating warranty data from service to design models such as CAD, Geometric Dimensioning & Tolerancing (GD&T) etc. Steps of the methodology include: (i) identification of faulty product features from the Ishikawa diagram of the failure reported in warranty; (ii) variation simulation analysis of geometric features of internal components; (iii) determination of critical geometric features affecting faulty product feature via surrogate modelling of dimensional variations; and (iv) analyzing sensitivity of faulty product features on critical geometric features. The proposed Design-Service inter loop is demonstrated by an industrial case study of automotive ignition switch and ‘Sticky Key’ service failure. RCA of ‘Sticky Key’ issue identifies critical geometric features and their sensitivity in affecting the faulty product features, which cause the failure