2-型模糊环境下的能源系统优化

基于模糊可能性理论,建立2-型模糊环境下的能源分配优化模型,其中各种类型能源的成本用2-型模糊变量刻画.用均值简约方法简约2-型模糊成本,建立广义期望值意义下的模糊能源分配优化模型.当成本用相互独立的三角2-型模糊变量刻画时,所建立的模糊能源分配优化模型可以转化为等价的参数线性规划.最后提供一个数值例子表明建模思想.     

Optimal trading strategy for European options with transaction costs

The European option with transaction costs is studied. The cost of making a transaction is taken to be proportional by a factor λ to the value (in dollars) of stock traded. When there are no transaction costs (i.e. when λ=0) the well-known Black-Scholes strategy tells how to hedge the option. Since no non-trivial perfect hedging strategy exists when λ>0 (see (Ann. Appl. Probab. 5(2) (1995) 327)), we instead try to maximize the expected utility attainable. We seek to understand the effect transaction costs have on the maximum attainable expected utility over all strategies, when λ is small but non-zero. It turns out that transaction costs diminish the expected utility by an amount which has the order of magnitude λ^2/3. We will compute that correction explicitly modulo an error which is small compared to λ^2/3. We will exhibit an explicit strategy whose expected utility differs from the maximum attainable expected utility by an error small in comparison to λ^2/3.     

Reinforcement learning for joint pricing,lead-time and scheduling decisions in make-to-order systems

The paper investigates a problem faced by a make-to-order (MTO) firm that has the ability to reject or accept orders, and set prices and lead-times to influence demands. Inventory holding costs for early completed orders, tardiness costs for late delivery orders, order rejection costs, manufacturing variable costs, and fixed costs are considered. In order to maximize the expected profits in an infinite planning horizon with stochastic demands, the firm needs to make decisions from the following aspects: which orders to accept or reject, the trade-off between price and lead-time, and the potential for increased demand against capacity constraints. We model the problem as a Semi-Markov Decision Problem (SMDP) and develop a reinforcement learning (RL) based Q-learning algorithm (QLA) for the problem. In addition, we build a discrete-event simulation model to validate the performance of the QLA, and compare the experimental results with two benchmark policies, the First-Come-First-Serve (FCFS) policy and a threshold heuristic policy. It is shown that the QLA outperforms the existing policies.     

The informational content of subordinated debt and equity prices in the presence of bankruptcy costs

The paper extends the contingent valuation framework of Black and Cox [J. Finance 31(2) (1976) 351] to value subordinated debt by explicitly incorporating bankruptcy costs in the model. We show that the information from subordinated debt prices is complementary to the information from the equity prices only when the bankruptcy costs are taken into account. In fact, the joint use of equity and subordinated debt prices can provide information on magnitude of expected bankruptcy costs. Knowing the magnitude of expected bankruptcy costs is necessary for calculating variables underlying policy objectives. In particular, it is illustrated that the value of expected liability of a deposit insurer would be underestimated if the bankruptcy costs were not taken into account.     

Optimal stochastic scheduling in a single server biclass retrial queueing system

The paper deals with the assignment of a single server to two retrial queues. Each customer reapplies for service after an exponentially distributed amount of time. The server operates at customer dependent exponential rates. There are holding costs and costs during service per customer and per unit of time. We provide conditions on which it is optimal to allocate the server to queue 1 or 2 in order to minimize the expected total costs until the system is cleared.     

A Flexible Parameterization for Baseline Mean Degree in Multiple-Network ERGMs

The conventional exponential family random graph model (ERGM) parameterization leads to a baseline density that is constant in graph order (i.e., number of nodes); this is potentially problematic when modeling multiple networks of varying order. Prior work has suggested a simple alternative that results in constant expected mean degree. Here, we extend this approach by suggesting another alternative parameterization that allows for flexible modeling of scenarios in which baseline expected degree scales as an arbitrary power of order. This parameterization is easily implemented by the inclusion of an edge count/log order statistic along with the traditional edge count statistic in the model specification.     

An optimal ordering policy for a spare unit with lead time

We consider a generalised ordering policy in which a spare unit for replacement can be delivered only by order after a constant lead time. Introducing the costs for ordering, shortage and holding, we derive the expected cost per unit time in the steady-state. We discuss the optimal ordering policy which minimises the expected cost, and show in a main theorem that the optimal policy is reduced to either one of two typical ordering policies depending on some conditions. We further discuss a similar ordering policy with varying lead times.     

Optimal control of a divergent multi-echelon inventory system

Consider a divergent multi-echelon inventory system, such as a distribution system or a production system. At every facility in the system orders are placed (or production is initiated) periodically. The order arrives after a fixed lead time. At the end of each period linear costs are incurred penalty costs are incurred at the most downstream facilities for back-orders. The objective is to minimize the expected holding and penalty costs per period. We prove that under the balance assumption it is cost optimal to control every facility by an order-up-to-policy. The optimal replenishment policy, i.e. the order-up-to-level and the allocation functions at each facility, can be determined by system decomposition. This decomposition reduces complex multi-dimensional control problems to simple one-dimensional problems.     

Integrated location and two-echelon inventory network design under uncertainty

In this paper, we propose a two-stage stochastic model to address the design of an integrated location and two-echelon inventory network under uncertainty. The central issue in this problem is to design and operate an effective and efficient multi-echelon supply chain distribution network and to minimize the expected system-wide cost of warehouse location, the allocation of warehouses to retailers, transportation, and two-echelon inventory over an infinite planning horizon. We structure this problem as a two-stage nonlinear discrete optimization problem. The first stage decides the warehouses to open and the second decides the warehouse-retailer assignments and two-echelon inventory replenishment strategies. Our modeling strategy incorporates various probable scenarios in the integrated multi-echelon supply chain distribution network design to identify solutions that minimize the first stage costs plus the expected second stage costs. The two-echelon inventory cost considerations result in a nonlinear objective which we linearize with an exponential number of variables. We solve the problem using column generation. Our computational study indicates that our approach can solve practical problems of moderate-size with up to twenty warehouse candidate locations, eighty retailers, and ten scenarios efficiently.     

A basis for modelling the costs of supplier selection: the economic tender quantity

In this paper, we consider the decision a purchaser must make regarding the number of suppliers that should be invited to submit a tender for a certain purchase. On the one hand, the costs of sending invitations to tender, the costs of evaluating the tenders received, and the costs of communicating the outcome to the suppliers that were not selected, all increase with the number of suppliers that are invited to tender. On the other hand, with every additional tender, the purchaser might obtain a better bid. We present a theoretical decision model, the Economic Tender Quantity (ETQ) model, that under certain conditions may assist the purchaser in finding the number of tenderers that minimises the expected total costs of the tender process, that is the sum of the expected costs of sending, evaluating and communicating, and the expected bid price. We discuss these conditions and possible extensions, as well as future research opportunities in this area.     

Economic optimization in a fixed sequence of unreliable inspections

Given a fixed sequence of unreliable inspection operations with known costs and inspection error probabilities of two types (classifying good items as defective and vice versa), we develop a model for selecting the set of inspections that should be activated in order to minimize expected total costs (inspection and penalties). We present an efficient branch and bound algorithm for finding the optimal solution, and two variations of a greedy heuristic that can be applied jointly to provide very good solutions at a O(n²) computational complexity. The conclusions are backed by a factorial experiment that included 1440 problem instances.     

An approximation algorithm for a facility location problem with stochastic demands and inventories

We propose a 2-approximation algorithm for a facility location problem with stochastic demands. At open facilities, inventory is kept such that arriving requests find a zero inventory with (at most) some pre-specified probability. Costs incurred are expected transportation costs, facility operating costs and inventory costs.     

基于存货质押的库存管理研究

针对基于存货质押的库存管理问题进行分析并构建模型,从还贷和不还贷两个角度,研究模型的最佳订货批量和质押量.模型以利润函数为目标函数,利润函数包括了销售额、缺货损失、剩余品价值、产品成本以及贷款本息和.通过对利润函数进行求解,得出最佳订购批量、存货质押量和最大期望利润,最后根据利润的大小决定是否偿还贷款.     

Maximum-revenue tariff under Bertrand duopoly with unknown costs

This paper considers an international trade under Bertrand model with differentiated products and with unknown production costs. The home government imposes a specific import tariff per unit of imports from the foreign firm. We prove that this tariff is decreasing in the expected production costs of the foreign firm and increasing in the production costs of the home firm. Furthermore, it is increasing in the degree of product substitutability. We also show that an increase in the tariff results in both firms increasing their prices, an increase in both expected sales and expected profits for the home firm, and a decrease in both expected sales and expected profits for the foreign firm.     

Applications of some formulas for finite Markov chains

We present some explicit matrix formulas for a finite state Markov chain. The first gives sums of probabilities along some general subsets of paths. Another formula yields the probability mass function (pmf) of the random variable which adds costs along subsets of paths. We then discuss how these formulas can be used to efficiently compute expected values of a function of the sum of costs along paths, as well as related applications. We conclude by describing a procedure allowing us to avoid using Monte Carlo simulation in stochastic approaches to solving some general boundary value problems. Instead, we show how to evaluate the relevant expected values exactly for discretizations of the original continuous problem.     

基于排序预测的带交易费用在线投资组合策略

在线投资组合决策过程中频繁调整资产头寸会产生较多的交易费用。本文提出了一个综合考虑预期收益和交易费用的在线投资组合策略。通过预测资产的排序计算组合的预期收益,利用相对熵距离衡量交易费用,构造了一个极大化预期收益和极小化交易费用的优化模型,从而得到了一个在线投资组合更新策略。然后,从理论上证明了该策略具有BH泛证券性,即该策略与离线的最优购买并持有策略具有相同的渐近平均指数收益率。最后,采用中美股票市场实际数据,对该策略进行了数值分析。结果表明,该策略的表现优于已有的在线投资组合策略,且对模型的参数不敏感。     

Optimal keyword bidding in search-based advertising with target exposure levels

Search-based advertising has become very popular since it provides advertisers the ability to attract potential customers with measurable returns. In this type of advertising, advertisers bid on keywords to have an impact on their ad’s placement, which in turn affects the response from potential customers. An advertiser must choose the right keywords and then bid correctly for each keyword in order to maximize the expected revenue or attain a certain level of exposure while keeping the daily costs in mind. In response to increasing need for analytical models that provide a guidance to advertisers, we construct and examine deterministic optimization models that minimize total expected advertising costs while satisfying a desired level of exposure. We investigate the relationship between our problem and the well-known continuous non-linear knapsack problem, and then solve the problem optimally by utilizing Karush–Kuhn–Tucker conditions. We present practical managerial insights based on the analysis of both a real-life data from a retailer and a hypothetical data.     

Optimal policy for a dynamic,non-stationary,stochastic inventory problem with capacity commitment

This paper studies a single-product, dynamic, non-stationary, stochastic inventory problem with capacity commitment, in which a buyer purchases a fixed capacity from a supplier at the beginning of a planning horizon and the buyer’s total cumulative order quantity over the planning horizon is constrained with the capacity. The objective of the buyer is to choose the capacity at the beginning of the planning horizon and the order quantity in each period to minimize the expected total cost over the planning horizon. We characterize the structure of the minimum sum of the expected ordering, storage and shortage costs in a period and thereafter and the optimal ordering policy for a given capacity. Based on the structure, we identify conditions under which a myopic ordering policy is optimal and derive an equation for the optimal capacity commitment. We then use the optimal capacity and the myopic ordering policy to evaluate the effect of the various parameters on the minimum expected total cost over the planning horizon.     

Leadtime management in a periodic-review inventory system: A state-dependent base-stock policy

We examine a single-item, periodic-review inventory system with stochastic leadtimes, in which a replenishment order is delivered immediately or one period later, depending probabilistically on costly effort. The objective is to determine a joint inventory policy and effort-choice strategy that minimizes the expected total costs. Our analytical and computational analysis suggests that (i) a state-dependent base-stock policy is optimal, (ii) the optimal effort strategy is such that the marginal cost of effort is equal to the value of immediate delivery, and (iii) the cost impact of leadtime reduction can be very large. We also provide some counter-intuitive results, compared with the traditional multi-period newsvendor model.     

An impulse control of a geometric Brownian motion with quadratic costs

We examine an optimal impulse control problem of a stochastic system whose state follows a geometric Brownian motion. We suppose that, when an agent intervenes in the system, it requires costs consisting of a quadratic form of the system state. Besides the intervention costs, running costs are continuously incurred to the system, and they are also of a quadratic form. Our objective is to find an optimal impulse control of minimizing the expected total discounted sum of the intervention costs and running costs incurred over the infinite time horizon. In order to solve this problem, we formulate it as a stochastic impulse control problem, which is approached via quasi-variational inequalities (QVI). Under a suitable set of sufficient conditions on the given problem parameters, we prove the existence of an optimal impulse control such that, whenever the system state reaches a certain level, the agent intervenes in the system. Consequently it instantaneously reduces to another level.