Integrated decision of release and patch planning for warrantied smart products considering hardware–software interaction

Smart products, such as smartphones, smart cars, and industrial robots, are increasingly integrated into daily life and industrial applications. To stay competitive, developers may release products without thorough testing, resulting in a certain number of faults remaining in the software, which may cause product failures during usage. To mitigate these risks and protect users, developers typically offer warranties that cover repair or replacement services for product failures. During the warranty period, they may also deploy software patches to fix identified faults, which may improve reliability but increase development cost. To balance product reliability with associated costs in competitive markets, this study proposes an integrated decision model of release and patch planning for warrantied smart products explicitly considering hardware–software interaction. The model enables developers to jointly optimize the release time, number of patches, and patch schedules by minimizing the total expected development cost. Both periodic and non-periodic patch scenarios are analyzed, with a dynamic programming algorithm developed to solve the non-periodic case. The applicability and performance of the proposed model are demonstrated through a numerical example using a real-world fault dataset. Numerical results show that the non-periodic patch scenario is not only more cost-effective but also facilitates earlier product releases than the periodic scenario.