Abstract

“Dynamic programming” (DP) is an influential tool for resolving an extensive class of sequential decision making problems under uncertainty. Technically, it enables us to compute optimal decision rules that spell out the best promising decision in any condition. This article reviews progresses in DP and distinguishes its ground breaking effects on economics, operations research, engineering and artificial intelligence (AI) with the relative scantiness of its real world appliances to upgrade the decision making of firms and individuals. The fuzziness of numerous real world decision problems and the difficulty in mathematically modeling them are main hurdles to an extensive application of DP in real world situations. Even so, in this review article a number of successful experiments were conferred, it was concluded that DP proposes significant assurance for better decision making. The empirically unsustainable hypothesis of unbounded rationality and tackle the challenging decision problems faced every day in routine life. Furthermore, it is deemed as a good technique for optimal operation due to the sequential decision making and ease in handling non-linear objective roles and constraints. Although the application of DP to major and multi directional situation is not that encouraging due to the problem ‘curse of dimensionality’. Incremental DP, discrete differential DP, DP with successive approximation and incremental DP with successive approximation are a few of the algorithms evolved to tackle this problem for DP. But in all these cases, it is difficult to choose an initial trial path, to obtain an optimal solution and there is uncontrolled number of iterations required for convergence.