Quality System for Production Software as Tool for Monitoring and Improving Organization KPIs

Abstract

In this paper we propose a solution as support for quality systems for production software. The motivation behind this study was to reduce that cost in the production area caused by gaps in the quality of the production software. Our proposal: QSPS (Quality System for Production Software) is offering support in the vulnerable points of these quality systems which usually generate nonconformities and have proved to be difficult or impossible to control. QSPS is a method in seven steps or modules that integrates also software tools, templates, checklists, evaluating tools elaborated complying to products, process and system quality standards. If other analyzed methods like: Scrum, XP, Fuzzy, Prompt, PTA, PRINCE2 or norms like: ISO 9001, ISO 9000-3, TickIT, CMM and CMMI, AQAP-110/AQAP-150, IEEE 730/983 are working in a reactive way, after the developing phase was finished, QSPS is an active system, helping the software developer from the beginning of the implementation phase to improve the developing methodology and to fulfill the quality requirements. QSPS model was applied in one of the largest European automotive company, the result being finalized in a practical approach of the QSPS, named QSMA − Quality System for Manufacturing Application. Using the QSMA (Quality System for the manufacturing application) for industrial projects and not only therefore, has led to accurate running of the production line from beginning of the SOP (Start of Production). Once this system was implemented and the production software applications were realized under the principles and rules of the QSMA, we defined strategically measurable KPIs (key performance indicators) out of the seven modules of the QSMA. This KPIs have the role to signal every time a production application has not the desired quality level and presents a high level of risk that could cause additional costs in the production. Based on the KPIs evolution, the weaknesses in the software applications can be identified in real time, so that the developer can react immediately, before occurrence of substantial damage.