隨著科技與技術的進步，產品推陳出新的速度也越來越快，因此汰換策略的制定也變得非常的重要。以及在現今工廠的大量製造與自動化，產品模組化及一體成形的情形越來越常見。當產品失效時，直接置換成一個新的產品，比起維修會更具有成本效益。一般而言，產品可以分為可維修產品與不可維修產品，本論文會將焦點放在不可維修之產品上，產品失效即置換，此過程符合更新過程(Renewal Process)，因此可以使用更新函數(Renewal Function)來模擬時間內產品需要更新的次數。並在新世代產品上市之後，開始考量汰換至新世代產品。新舊世代產品會有系統不相容的情況發生，所以在決定汰換之後，會有汰換成本的產生。本論文主要探討不同壽命分配之不可維修產品汰換至新世代產品之數學模型，並在最後進行數值分析尋求最佳汰換時間，並判斷成本參數對最佳汰換時間的影響。

With the development of sciences and technology, the cycle of new-generation product releases is getting shorter. So how to make the decision on the switch-over policy become very important. And now the factories are produce the product in mass production and automation way. The all-in-one product become more and more common. When a product fails, it is more cost-effective to replace it with a new one than to repair it. According to the characteristic of the product, we can divide it into repairable products and non-repairable products. This thesis will focus on non-repairable products. When the products fail, we will perform replacement directly to replace with new product. This process complies with the renewal process, so we can use renewal function to simulate the frequency of the product failed during this time period. After the launch of the new generation products, customer may consider to switch-over the old generation products to the new generation products. There will be system incompatibility between new and old generation products, so after to do switch-over, there will be switch-over costs. This thesis mainly discusses the model of switch-over to the new generation product with different lifetime distribution. Finally do the numerical analysis to find the optimal switch-over time and find the impact of cost parameters on the optimal switch-over time.

[1] 張嘉文，「汰換為新世代產品之最佳時間」，國立台灣科技大學工業管理系碩士論文 2019。
[2] 陳柏廷，「新世代購買成本與汰換成本隨時間變動下產品之最佳汰換時間」，國立台灣科技大學工業管理系碩士論文 2020。
[3] 吳嘉宥，「即將停產產品之最終單次訂購量及最佳汰換時機」，國立台灣科技大學工業管理系碩士論文 2021。
[4] 黃允成、謝雅惠、邱士珍，「多類型故障型態下最適預防保養策略研究，」技術學刊，卷 32，頁碼 13-38，2017。
[5] 黃允成、李柏霆、陳怡妗，「在給定壽命分配下最適設備預防保養策略之研究」品質學報，卷 23，頁碼 309-332，2016。
[6] Chang, C. C., Sheu, S. H., and Chen, Y. L., “Optimal number of minimal repairs before replacement based on a cumulative repair-cost limit policy,” Computers & Industrial Engineering, 59(4), 603-610 (2010).
[7] Smith, W, and Leadbetter, M., “On the renewal function for the Weibull Distribution,” Technometrics, 5(3), 393-396 (1963).
[8] Smeitink, E., and Dekker, R., “A simple approximation to the renewal function,” IEEE Transactions On Reliability, 39(1), 71-75 (1990).
[9] Kumar, U., and Knezevic, J., “Availability based spare optimization using renewal process,” Reliability Engineering and System Safety, 59(2), 217-223 (1998).
[10] Jiang, X., Makis, V., and Jardine, A. K., “Optimal repair/replacement policy for a general repair model,” Advances in Applied Probability, 33(1), 206-222 (2001).
[11] Yeh, R. H., Kurniati, N., Chang, W. L., “Optimal Single Replacement Policy for Products with Free-Repair Warranty Under a Finite Planning Horizon, ” Quality Technology & Quantitative Management, vol. 12, pp. 159-167, 2015.
[12] Park, M., and Pham, H., “Cost models for age replacement policies and block replacement policies under warranty, ” Applied Mathematical Modelling, vol. 40, pp. 5689-5702, 2016.
[13] Kusaka, Y., and Suzuki, H., “Equipment replacement behavior under innovative technological advances,” Journal of the Operations Research Society of Japan, 33(1), 76-99 (1990).
[14] Bylka, S., Sethi, S., and Sorger, G., “Minimal forecast horizons in equipment replacement models with multiple technologies and general switching costs,” Naval Research Logistics, 39(4), 487-507 (1992).
[15] Chand, S., McClurg, T., and Ward, J., “A single‐machine replacement model with learning,” Naval Research Logistics, 40(2), 175-192 (1993).
[16] Bethuyne, G., “Optimal replacement under variable intensity of utilization and technological progress,” The Engineering Economist, 43(2), 85-105 (1998).
[17] Yatsenko, Y., and Hritonenko, N., “Discrete–continuous analysis of optimal equipment replacement,” International Transactions in Operational Research, 17(5), 577-593 (2010).
[18] Constantine, A.G., and Robinson, N.I., “The Weibull renewal function for moderate to large arguments,” Computational Statistics & Data Analysis, 24(1), 9-27 (1997).