The Optimal Car Replacement Strategy: However, the reality is more complex
2025 (English)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE credits
Student thesis
Abstract [en]
Every car owner makes a decision each day, implicit or explicit, on whether to keep their current car or sell it and possibly replace it with another. This thesis explores the optimal decision-making strategy, referred to as a policy, for minimizing car ownership costs. The optimal car replacement policy is derived using a car replacement model introduced in the 1960s, with a methodology based on an algorithm called policy iteration.
Through policy iteration, the study identifies the optimal cost-minimizing strategy: purchasing a car aged 28 years and keeping it until the end of its lifespan. However, recognizing the practical challenges of buying cars nearing the end of their lifespan, the thesis also evaluates a scenario where restrictions are imposed, limiting purchases to cars 19 years old or younger. Under such constraints, the optimal policy shifts to buying a car aged 16 years and retaining it until the end of its lifespan.
While this thesis provides valuable insights into cost-effective car replacement strategies, it acknowledges that the car replacement model introduced in this thesis may be an overly simplified representation of the real world. Incorporating additional factors, such as safety considerations or personal preferences, could enhance the model and make it more applicable to practical decisionmaking.
Place, publisher, year, edition, pages
2025. , p. 36
Keywords [en]
Reinforcement learning, Dynamic programming, Policy iteration, Markov decision process, Optimal policy, Car replacement
National Category
Mathematics
Identifiers
URN: urn:nbn:se:lnu:diva-134978OAI: oai:DiVA.org:lnu-134978DiVA, id: diva2:1932263
Subject / course
Mathematics
Educational program
Applied Mahtematics Programme, 180 credits
Supervisors
Examiners
2025-01-312025-01-282025-01-31Bibliographically approved