Self-healing has been increasingly integrated into systems to enhance their reliability. Many popular intrinsic self-healing policies are not cost-effective because their self-healing actions are not always performed ...
详细信息
Self-healing has been increasingly integrated into systems to enhance their reliability. Many popular intrinsic self-healing policies are not cost-effective because their self-healing actions are not always performed at the right time. This paper investigates the economic design of a self-healing policy with limited agents for an intelligent system that executes a mission of finite length. Several healing agents are embedded into the system before the mission. The system performs self-detection at equidistant epochs to reveal deterioration levels. The deterioration can be randomly healed by releasing healing agents, and the healing effect depends on the number of agents released. At each inspection epoch, a decision is made on how many healing agents to be released. The objective is to jointly determine the capacity of agents and the self-healing policy to minimize the expected total cost over the mission length. The optimization problem is formulated in the stochastic dynamic programming framework. A backward induction algorithm is developed to find the optimal solution. A numerical example is provided to illustrate the effectiveness of the proposed approach. The comparison with a control-limit policy confirms the outstanding performance of the proposed policy.
Insect pests pose a threat to humans by jeopardizing food security in agricultural systems, acting as vectors for infectious diseases, and damaging forests and other ecosystems. Despite decades of research, effective ...
详细信息
Insect pests pose a threat to humans by jeopardizing food security in agricultural systems, acting as vectors for infectious diseases, and damaging forests and other ecosystems. Despite decades of research, effective pest management remains challenging. Incomplete understanding of the mechanisms behind pest population dynamics limits our ability to anticipate outbreaks. Hence, pest management is often reactive, meaning control actions are taken once outbreaks have already begun, allowing for damage to occur. Here we show that a datadriven model can effectively predict outbreaks, allowing us to optimize control strategies, targeting pests before outbreaks occur. Specifically, we explore empirical dynamic modeling paired with stochastic dynamic programming to keep insect populations within acceptable bounds. We show that this framework reduces outbreaks in several simulated and empirical scenarios. Our study provides a promising framework to reduce losses from pests.
Cross-training is an effective way to acquire labor flexibility which can be defined by multifunctionality and redundancy. This study concentrates on redundancy and considers how much capacity is required to satisfy p...
详细信息
Cross-training is an effective way to acquire labor flexibility which can be defined by multifunctionality and redundancy. This study concentrates on redundancy and considers how much capacity is required to satisfy present and future demand. A two-phase cross-training problem is investigated to discuss how much redundant capacity should be pooled for the bottleneck work station of a new product with the consideration of demand uncertainty. The problem determines the number of workers should be trained in the first phase and in which period they should fulfill the second phase. Workers participate in the first phase should be more than the current demand needed. These extra workers act as a capacity buffer against demand uncertainty, and training costs paid for them can be treated as an investment in flexibility. The decision process is analogous to an expansion option, and a real options approach is employed to solve the problem. A stochastic dynamic programming model, in which the dynamic of demand is approximated by a trinomial lattice, is formulated. Based on differences in the expected NPV brought by training extra workers, the monetary value of labor flexibility can be measured. Then, the impact of demand uncertainty, workers' multi-skill level and penalty cost for unsatisfied demand on the optimal training plan and the value of flexibility are tested by using numerical experiments. The results verify that flexibility brings more value when demand variation is great and pooling redundant capacity in advance is necessary to cope with demand uncertainty. Besides, the results have several implications. (1) In order to rise profits, a lower multi-skill level is suggested for products with great demand variation. (2) The quantity of capacity pooled for the bottleneck is crucial to the value of flexibility. (3) Excessive multifunctionality hurts the value of flexibility brought by redundancy. (4) Flexibility brought by capacity buffer is valuable for companies who a
This study presents a stochastic dynamic programming-based cooling controller for the battery thermal management system (BTMS) in electric vehicles (EVs). Addressing the complex interplay between battery performance a...
详细信息
This study presents a stochastic dynamic programming-based cooling controller for the battery thermal management system (BTMS) in electric vehicles (EVs). Addressing the complex interplay between battery performance and safety, our approach optimizes temperature regulation while minimizing power consumption. Notable contributions include minimum transitions between refrigeration and radiator modes, integration of an artificial neural network for computing efficiency, and an infinity-horizon expected cost formulation considering future heat disturbances. Comparative analyses demonstrate superior performance, showcasing the proposed controller's efficiency in achieving smaller battery temperature variation and consistently lower energy consumption across diverse ambient conditions. The proposed controller shows 56% lower energy consumption on average compared to rule-based controller.
Many sequential decision problems involve deciding how to allocate shared resources across a set of independent systems at each point in time. A classic example is the restless bandit problem, in which a budget constr...
详细信息
Many sequential decision problems involve deciding how to allocate shared resources across a set of independent systems at each point in time. A classic example is the restless bandit problem, in which a budget constraint limits the selection of arms. Fluid relaxations provide a natural approximation technique for this broad class of problems. A recent stream of research has established strong performance guarantees for feasible policies based on fluid relaxations. In this paper, we generalize and improve these recent performance results. First, we provide easy-to-implement feasible fluid policies that achieve root ffiffififfi performance within O( N ) of optimal, where N is the number of subproblems. This result holds for a general class of dynamic resource allocation problems with heterogeneous subproblems and multiple shared resource constraints. Second, we show using a novel proof technique that a feasible fluid policy that chooses actions using a reoptimized fluid value root ffiffififfi function achieves performance within O( N ) of optimal as well. To the best of our knowledge, this performance guarantee is the first one for reoptimization for the general dynamic resource allocation problems that we consider. The scaling of the constants with respect to time in these results implies similar results in the infinite horizon setting. Finally, we develop and analyze a class of feasible fluid-budget balancing policies that stay "close" to actions selected by an optimal fluid policy while simultaneously using as much of the shared resources as possible. We show that this policy achieves performance within O(1) of optimal under particular nondegeneracy assumptions. This result generalizes recent advances for restless bandit problems by considering (a) any finite number of actions for each subproblem and (b) heterogeneous subproblems with a fixed number of types. We demonstrate the use of these techniques on dynamic multiwarehouse inventory problems and find empirical
The annual compliance cycle of the carbon trading system gives generation companies (GenCos) the flexibility to buy allowances at preferred times rather than the same time that emissions happen. How to estimate the ac...
详细信息
The annual compliance cycle of the carbon trading system gives generation companies (GenCos) the flexibility to buy allowances at preferred times rather than the same time that emissions happen. How to estimate the actual emission cost under fluctuating carbon prices and make better generation decisions become challenges for GenCos. A synergistic generation and carbon trading optimization method is proposed in this paper with a nonspeculative dynamic (NSD) carbon trading strategy. The NSD strategy utilizes stochastic dynamic programming to make optimal carbon trading decisions and to provide accurate emission cost estimations for GenCos to make generation decisions. To control risks, the NSD strategy forbids intentional speculations in the carbon market but allows both buying and selling of allowances according to the yearly total emissions estimated by the generation strategy. A risk control scheme by adjusting the allowance-emission balancing period is proposed so that GenCos can smoothly shift between the traditional daily-balanced carbon trading strategy and the NSD strategy. Properties of the NSD strategy are analyzed and proved theoretically under some sufficient conditions, which makes the strategy interpretable and trustworthy. Numerical simulations verify that the proposed method can help GenCos lower their emission costs and improve their overall profits in a year.
The zero-emission zone (ZEZ) is a recent environmental regulation that restricts the entry of internal combustion engine vehicles. In a ZEZ, hybrid electric vehicles (HEVs) are allowed but must operate in full-electri...
详细信息
The zero-emission zone (ZEZ) is a recent environmental regulation that restricts the entry of internal combustion engine vehicles. In a ZEZ, hybrid electric vehicles (HEVs) are allowed but must operate in full-electric mode. Therefore, it is important for HEVs entering a ZEZ to have a sufficiently charged battery. This study presents a stochastic dynamic programming-based power management strategy for optimizing HEV charging in preparation for ZEZ drives. stochastic dynamic programming models the driver's intentions as a Markov chain and designs optimal controllers by incorporating future probabilistic information up to an infinite time horizon. Furthermore, the proposed controller takes into account the remaining distance to the zero-emission zone, enabling efficient charging. Compared to stochastic dynamic programming strategies that do not consider the remaining distance, the proposed power management strategy improves the equivalent fuel efficiency by up to about 21%.
Some nonprofit organizations (NPOs) manage a complex workforce composed of a mix of volunteers, part-time workers, and full-time workers. We study the NPO's finite-horizon staffing problem to determine the optimal...
详细信息
Some nonprofit organizations (NPOs) manage a complex workforce composed of a mix of volunteers, part-time workers, and full-time workers. We study the NPO's finite-horizon staffing problem to determine the optimal initial staff planning decisions and per period optimal hiring and assignment decisions given a budget, capacity constraints, and an uncertain supply of volunteers and part-time workers. Our main goal is to solve this problem in a way that is effective and easy to implement while obtaining interesting managerial insights. To this end, we first demonstrate that the optimal staffing policies are computationally challenging to identify in general. However, we demonstrate that a prioritization assignment policy and a hire-up-to policy for part-time workers can be conveniently applied and are close to optimal. These policies are, in fact, optimal under staff scarcity and staff sufficiency. In our numerical analysis, we study the value and impact of the general optimal solution that considers flexibility and turnover of part-time workers versus the prioritization assignment policy and a constant hire-up-to policy that omit flexibility and turnover behaviors. We further suggest two easy-to-implement heuristics and theoretically analyze them and run a numerical performance study. We observe that both heuristics have low relative optimality gaps. Finally, we extend our analysis by studying how the optimal policy varies under three different practical considerations: a concave social value objective, nonzero volunteer costs, and dynamic volunteer behaviors.
Consider a newly developed system sold under a performance based contract (PBC), and subject to failures following various failure modes. Redesign may address certain failure modes. To trade-off redesign costs and cos...
详细信息
Consider a newly developed system sold under a performance based contract (PBC), and subject to failures following various failure modes. Redesign may address certain failure modes. To trade-off redesign costs and costs associated with the PBC, we find the optimal failure modes to redesign based on the current beliefs on the failure rate per failure mode, and the amount of time remaining under the PBC. We develop analytic insights into the structure of the optimal policy. (c) 2024 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons .org /licenses /by /4 .0/).
In this paper, we study the optimal management of a target benefit pension plan. The fund manager adjusts the benefit to guarantee the plan stability. The fund can be invested in a riskless asset and several risky ass...
详细信息
In this paper, we study the optimal management of a target benefit pension plan. The fund manager adjusts the benefit to guarantee the plan stability. The fund can be invested in a riskless asset and several risky assets, where the uncertainty comes from Brownian and Poisson processes. The aim of the manager is to maximize the expected discounted utility of the benefit and the terminal fund wealth. A stochastic control problem is considered and solved by the programmingdynamic approach. Optimal benefit and investment strategies are analytically found and analyzed, both in finite and infinite horizons. A numerical illustration shows the effect of some parameters on the optimal strategies and the fund wealth.
暂无评论