Assessing the Release Readiness of Engine Control Software

Abstract

Powertrain control calibration is an essential stage before the delivery of final products, to ensure that a vehicle works well in all driving environments. However, due to complexity of a powertrain control system (often as many as 40,000 parameters), it is difficult to assess when all parameters are fully calibrated. Therefore, the aim of this paper is to explore an approach to evaluate and predict the maturity level of powertrain control software based on calibration data for simulation models. We developed metrics that indicate software maturity and their visualization using heatmaps. We used software maturity growth curves to select maturity growth models for maturity assessment. The results and approaches of this paper were validated with theoretical methods and empirical methods using action research techniques. Our conclusions show that we can use standard software reliability growth models to monitor the calibration process for powertrain software, and we conclude that this type of monitoring provides a good support for quality stakeholders in monitoring powertrain calibration quality.