Warranties play an important role for marketing products in the intensly competitive market, since an attractive policy can dramatically increase market share. However, offering an unlimited warranty with intent to monopolize the market is unrealistic because of the extremely high cost for maintaining such a policy. This paper considers a decision problem under the policy of a pro-rata warranty (PRW), and a Bayesian decision model in determining the optimal warranty proportion is proposed, where a periodic preventive maintenance (PM) program is performed during the warranty term to slow down the product deterioration, and a nonhomogeneous Poisson process is employed to describe the successive breakdown times of the deteriorating product. The proposed approach provides decision-making techniques, not only for taking action in the light of all available relevant information, but also for maximizing expected profits. Finally, an application case is utilized to demonstrate the feasibility and validity of the proposed approach, and sensitivity analyses regarding the important factors that may have impacts on the warranty proportion are also discussed.Index Terms-Bayesian analysis, deterioration, minimal repair, nonhomogeneous poisson process (NHPP), power law intensity.