本文延伸Nair(1989)之技術性報廢與馬可夫耗棋下設備重直模式，假設囡科技新知識的街擊累積，導致單一技術突破出現，促使較優性能之改良型設備上市o因此，是否續用現行設備或購置新型設備，以求取無限時間幅度下之最大總期望折現報酬，將成為管理上之重要課題o有鑒於此，本研究之最佳政策，同時考量報酬函數與技術突破條件機率分佈函數。其次，設若報酬函數不僅為設備狀態，亦為使用年m之函數;有關技術突破條件機率分佈函數之建立與推證'則係沿修Aven與Garrder(1987)之衝 擊模式而得。據此，引用隨機控制理論，建構最佳值函數之動態遞迴式o在合理假設條件下，證明技街預測時期存在，且自該時期以後，技街突破條件機率為常數;而最佳重置政策結構，可以設備狀態與使用年限組成之特殊控制極限表之o再者，若技術突破機率為時間之非遞減函數，則最佳值函數亦為時間之非遞減函數o最後，證明最佳重置政策存在預測幅度O

Extending Nair (1989) study， this paper deals with the equipment replacement under Markovian deterioration and technological obsolescence. We assume that the accumulated impact of the new knowledge results in a single technology breakthrough and consequently improved equipment is available on the market. A crucial issue in management is to decide for each period whether to keep using the existing equipment or to procure new equipment so as to maximize the total expected discounted reward in the infinite horizon. Two functions have to be taken into consideration for deciding on an optimal policy， namely， reward functions and conditional probability distribution function of technology breakthrough. The reward functions depend on the state and the age of equipment， and the conditional probability distribution of technology breakthrough is based on the shock model of Aven and Gaarder (1987) with revision. We then establish the recursive formula by using stochastic control theory. We prove that the conditional probability distribution of technology breakthrough is constant after technology forecast period under some reasonable conditions. We also show that the optimal policy has a special control limit structure form of equipment state and age. Furthermore， we prove that the optimal value function is nondecreasing when technology breakthrough rates are nondecreasing in time. Ultimately，the existence of a forecast horizon for optimal replacement policy is demonstrated.