Stochastic Demand Model Research Articles

High-speed rail transport valuation

In this paper, the optimal timing for investing in high-speed rail projects under uncertainty in relation to the utility provided to railway users was investigated. To accomplish this, a continuous time real options analysis framework using a stochastic demand model was developed to determine the optimal time to invest. Uncertainty upon investment expenditures was also added in an extended framework. The value of the option to defer and the investment opportunity value were also assessed.

Stochastic Demand Vehicle Routing Problem on Immune and Genetic Algorithm

In this paper, stochastic demand vehicle routing problem and model are studied. Hybrid algorithm based on immune algorithm and genetic algorithm is proposed in order to gain solution. IA-GA use basic principles and processes of genetic algorithm, and includes the promotion-inhibition function and memory function of immune algorithm. The premature convergence problem of genetic algorithm is conquered, and we can get different offspring with improved crossover. Comparing experiment shows that IA-GA has better computing performance.

Stochastic Demand Vehicle Routing Problem on Immune and Genetic Algorithm

In this paper, stochastic demand vehicle routing problem and model are studied. Hybrid algorithm based on immune algorithm and genetic algorithm is proposed in order to gain solution. IA-GA use basic principles and processes of genetic algorithm, and includes the promotion-inhibition function and memory function of immune algorithm. The premature convergence problem of genetic algorithm is conquered, and we can get different offspring with improved crossover. Comparing experiment shows that IA-GA has better computing performance.

Game theoretic pricing and ordering decisions with partial lost sales in two-stage supply chains

Few papers have studied the optimal pricing and ordering with partial lost sales from a two-stage game theoretic perspective. This paper considers a supply chain game with a buyer and a seller through pricing and ordering models with partial lost sales. We evaluate the optimal solutions for the buyer and the seller and provide insights into both the deterministic and stochastic demand models. We also propose search algorithms to obtain the global optimum when convexity and concavity conditions are not satisfied.

High Speed Rail Transport Valuation

In this paper, the optimal timing for investing in high speed rail projects under uncertainty in relation to the utility provided to railway users was investigated. To accomplish this, a continuous time real options analysis framework using a stochastic demand model was developed to determine the optimal time to invest. The value of the option to defer and the investment opportunity value were also assessed.

Ant Colony Optimization for Solving Solid Waste Collection Scheduling Problems

Problem statement: Southern Waste Management environment (SWM environment) is a company responsible for the collection and disposal of solid waste for the city of Johor Bahru, a city with over one million populations. The company is implementing an integrated solid waste management system where it involved in the optimization of resources to ensure the effectiveness of its services. Formulating this real life problem into vehicle routing problem with stochastic demand model and using some designed algorithms to minimize operation cost of solid waste management. Approach: The implementation of Ant Colony Optimization (ACO) for solving solid waste collection problem as a VRPSD model was described. A set of data modified from the well known 50 customers problems were used to find the route such that the expected traveling cost was minimized. The total cost was minimized by adopting a preventive restocking policy which was trading off the extra cost of returning to depot after a stock-out with the cost of returning depot for restocking before a stock-out actually occurs. For comparison purposes, Simulated Annealing (SA) was used to generate the solution under the same condition. Results: For the problem size with 12 customers with vehicle capacity 10 units, both algorithms obtained the same best cost which is 69.4358 units. But the percentage deviations of averages from the associated best cost are 0.1322 and 0.7064 for ACS and SA. The results indicated that for all demand ranges, proposed ACO algorithm showed better performance than SA algorithm. Conclusion: SA was able to obtain good solutions for small ranges especially small size of problem. For ACS, it is always provide good results for all tested ranges and problems sizes of the tested problem.

Impacts of water cycle changes on the rice market in Cambodia: stochastic supply and demand model analysis

A supply and demand model for rice in Cambodia, which includes among other factors evapotranspiration as a water supply variable impacting regional yields and planted areas, is developed to aid in the design of agricultural policies and planning. Impacts are determined stochastically by drawing on water cycle distributions and evaluating the resulting variation in production and price bands for local rice markets. The results of the baseline analyses indicate that production of wet and dry season rice steadily increases and the consumption per capita slightly decreases due to the negative income elasticity. Results of a partial stochastic analyses show that the production of rice in regions where elevations are high and the land vulnerable to flooding are the most sensitive to increased fluctuations in water supply. The changes also affect the rice market through equilibrium price changes. The upper price band, which is the width between average and 90th percentile, is larger than the lower band, which is the width between average and tenth percentile, suggesting that the situation of low income consumers could grow worse under an unstable environment with relatively larger upward price spikes. The results imply that development of irrigation facilities and water management systems maybe required for Cambodian provinces which rely heavily on agriculture, particularly rice production, under increasing climatic variation.

Short-term shift setting and manpower supplying under stochastic demands for air cargo terminals

A good air cargo terminal manpower supply plan helps terminals deal efficiently with their cargos and reduces their operating costs. To design a good air cargo terminal manpower supply plan, a terminal has to consider not only its operating costs, but also the uncertainty of the manpower demand in actual operations. However, most air cargo terminals in Taiwan currently depend on staff experience with a fixed demand when establishing the manpower supply plan, which is neither effective nor efficient. We have developed two stochastic-demand manpower supply plan models for air cargo terminals that can resolve stochastic demands occurring in practice. The objectives of both models are to minimize the total man-hour cost, subject to the related operating constraints. The models are formulated as integer/mixed integer linear programs. To evaluate the two stochastic-demand models under stochastic demands, we have also developed two deterministic-demand manpower supply plan models, by suitably modifying two stochastic-demand models, respectively, and an evaluation method. Here, we perform a case study using real operating data from a Taiwan air cargo terminal. The preliminary results are good, showing that the models could be useful for planning air cargo terminal manpower supply.

Specification and Identification of Stochastic Demand Models

This paper is concerned with stochastic demand systems for continuous choices that arise from structural random utility models. It examines under which nonparametric conditions on the structural random utility specification the implied reduced form model is nonsingular and invertible. For parametric members within this class of random utility models, the paper provides conditions for local identification from the reduced form under moment assumptions.

Solving stochastic demand inventory models with uncertain data

We study the paper of Levy published in the journal of Management Sciellce. His paper discussed the cost influences for performing inventory policy calculated from inaccurate information. We point out that several important improvements can be achieved to refine his results. The purposes of this paper is threefold. First, we show that the Levy’s result on the proportionate augment in price can be further simplified. Second, there are symmetric relations in the stochastic demand model. Third, two results from the sensitivity analysis are presented.

Towards the development of an intelligent inventory management system

This paper is concerned with the development of an intelligent inventory management system which aims at bridging the substantial gap between the theory and the practice of inventory management. The proposed system attempts to achieve this by providing automatic demand and lead time pattern identification and model selection facilities. The process of demand pattern identification together with the statistical tests used is discussed. The models incorporated cover deterministic demand models including: constant, quasi‐constant, trended and seasonal demand as well as stochastic demand models. This paper includes an empirical evaluation of the system on real data from the manufacturing and airline industries which shows that this system can lead to significant savings in inventory cost.

Lot scheduling problem for continuous demand

This paper deals with a lot scheduling problem for multi-item continuous demand utilizing a single machine with setup times. The problem involves dynamic determination of lot-sizes to minimize the inventory level. In such systems, the lowest inventory level is realized by maximizing the number of setups under the condition of no stockout. Based on the above fact, first, an analytical solution of lot-sizes for a cyclic schedule is given based on a constant-demand model. Secondly, a stochastic-demand model is developed, and some heuristic setup rules are studied using the concept of time inventory, defined from the solution based on a constant-demand model. These rules realize a stable system with a lower average stockout level and a lower inventory level. Effects of the setup rules on the system are revealed by Monte Carlo simulations.