Dynamic Programming and Statistical Modelling in Automatic Speech Recognition

This article describes the methods which form the basis of contemporary automatic speech recognition systems. The two most prominent algorithms, dynamic time-warping and hidden Markov modelling, are described and compared. Particular attention is given to the role of dynamic programming in either approach.     

目标规划在离散型动态投入产出模型中的应用

以离散型动态投入产出模型为约束条件的主体、以决策部门所希望达到的种种目标为约束条件的附加部分,建立动态投入产出目标规划模型.通过求解动态投入产出目标规划模型而得到离散型动态投入产出模型的解.此解法与其它解法相比具有更大的实用价值.     

An Introduction to Dynamic Programming: The Theory of Multi-Stage Decision Processes

Journal of the Operational Research Society -     

Converging Marriage in Honey-Bees Optimization and Application to Stochastic Dynamic Programming

In this paper, we first refine a recently proposed metaheuristic called “Marriage in Honey-Bees Optimization” (MBO) for solving combinatorial optimization problems with some modifications to formally show that MBO converges to the global optimum value. We then adapt MBO into an algorithm called “Honey-Bees Policy Iteration” (HBPI) for solving infinite horizon-discounted cost stochastic dynamic programming problems and show that HBPI also converges to the optimal value.     

动态规划在教学管理中的应用

本将动态规划理论应用于高校教学管理当中，得出了教学管理中重要的环一排课的最优策略。     

An Example of Problem Embedding in Deterministic Dynamic Programming

A criticism sometimes made of dynamic programming is that in deterministic problems, optimal decisions are calculated which are never needed, as the decisions relate to states which never arise. In this paper we describe how some of these "redundant" calculations have been used, in a certain problem, to derive a working rule of general validity.The problem concerns the installation and replacement of an item of electrical equipment at a large number of sites, to meet a known load at each site which increases exponentially at a known rate.     

Extension of Dynamic Programming to Multi-plant Renewal

Discounting formulae and dynamic programming methods have beenproposed for single machine replacement problems with and withouttechnological change. But it is often essential to considercapacity replacement collectively in terms of all the plantsowned by a company as the replacement decisions are not independent.This paper proposes a model for capacity renewal and describeshow the problem can be tackled by dynamic programming combinedwith a priority ordering system.     

动态模糊规划模型的构建及应用

常规规划模型通常存在如下两种缺陷:首先,它的目标系数及约束条件都是在硬性限制下的确定值,因而在建模方面弹性小、硬度大;其次,它的目标系数与时间无关,因此不能有效地刻划时时刻刻变化着的目标系数,而动态模糊规划模型可以有效地解决上述缺陷.首先应用模糊动态AHP确定目标系数;然后根据L-R模糊数的强序关系准则,将动态模糊规划模型分解为最优与最劣两个模糊规划模型;再根据以α水平截集为基础的求解方法,将上述两个模型进行相应的转换,建立具有风险分析功能的动态模糊规划模型;最后将其应用到一个实际算例中,收到较好的结果.     

An Efficient Policy Iteration Algorithm for Dynamic Programming Equations

We present a scheme for Hamilton-Jacobi-Bellman equations based on a semi-Lagrangian discretization and an iterative method in the policy space. The scheme exploits the idea that a good initialization of the policy iteration procedure yields a faster numerical convergence to the optimal solution. The scheme features a pre-processing step with value iterations on a coarse grid. Numerical tests assess the efficient performance of the method. (© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim)     

基于故障诊断专家数据库系统的模糊模式识别

本文分析了模糊逻辑方法进行诊断的不足,提出了基于广义贴近度的模糊模式识别方法,建立并讨论了相应的故障诊断专家数据库系统     

Dynamic Programming Applied to Ship Fleet Management

A ship fleet management problem is described which concerns the phasing out of a fleet of general cargo ships over a 10-year period with the possibility of premature sale and temporary replacement by charter ships. The objective is to determine a sale and replacement policy which maximizes long term company assets. The problem is tackled by selecting a priority order for sale and determining the optimal rate of replacement by dynamic programming. The method has been programmed and demonstrated good returns and is in current use by a shipping company for a variety of management functions.     

An Approach to Optimal Recruitment and Transition Strategies for Manpower Systems using Dynamic Programming

In this paper, concepts of dynamic programming are used within a discrete time Markovian model for the development of a graded population. Optimal recruitment and transition patterns are determined by minimizing expected discrepancies between actual states and preferred goals.     

Optimal Battery Aging: An Adaptive Weights Dynamic Programming Algorithm

We present an algorithm to handle the optimization over a long horizon of an electric microgrid including a battery energy storage system. While the battery is an important and costly component of the microgrid, its aging process is often not taken into account by the energy management system, mostly because of modeling and computing challenges. We address the computing aspect by a new approach combining dynamic programming, decomposition and relaxation techniques. We illustrate this adaptive weight’ method with numerical simulations for a toy microgrid model. Compared to a straightforward resolution by dynamic programming, our algorithm decreases the computing time by more than one order of magnitude, can be parallelized, and allows for online implementations. We believe that this approach can be used for other applications presenting fast and slow variables.     

Fuzzy Pattern Recognition Approach to Construction Contractor Selection

Contractor selection is a complex process crucial to ensuring the success of construction projects. Existing methods by which owners select a suitable contractor have been inadequate because it is difficult for decision-makers to evaluate contractor bids against inexact qualitative criteria. The purpose of this paper is to propose a Multiple-layer Fuzzy Pattern Recognition (MFPR) approach to solve contractor selection problem. Integrating judgments, experience and preferences of decision-makers, this paper uses the paired comparison method to decide relative membership degrees of qualitative criteria as well as weights of the criteria set, a case study for a channel construction project was used to demonstrate the feasibility of this approach. The findings clearly indicate that the techniques may successfully harmonize different opinions and reach a group consensus.     

一类动态规划的解法及其在Turbo码解码中的应用

研究一类多阶段动态规划问题,给出了求最优解的方法.将这种多阶段动态规划应用于Turbo译码中,不仅可以减少运算量,还可以避免传统的Turbo译码算法需要进行指数运算以及其随着迭代次数的增加容易出现的数据溢出问题,因此是一种十分有效的方法,是对系统工程理论应用领域的拓宽.     

A Non-Serial Dynamic Programming Model with an Application to the Optimal Design of Distribution Networks

A non-serial dynamic programming (NSDP) formulation, aimed at the decomposition and optimization of interconnected systems, is presented. A new model is proposed for the system decomposition. It is used to optimize an electrical distribution system, and the results on a 25-node network are reported.