Asia-Pacific Journal of Operational Research

In this research, we discuss three different approaches to generate demand forecasting and pricing decision for mix of national brand and store brand products in the era of big data. We derive the equilibrium wholesale price and retail price for the national brand products, and the equilibrium retail price for the store brand products based on demand forecast under three different information scenarios, including Noninformation Sharing ($N$), Information Sharing ($I$), and Retailer Forecasting ($R$). We comprehensively discuss how information collection, information sharing, forecast accuracy under era of big data affect firms’ prices and profits. Our numerical experiments illustrate and verify our analytical findings and provide further managerial insights and interpretations.

With the rapid development of the Internet and big data technologies, a rich of online data (including news releases) can helpfully facilitate forecasting oil price trends. Accordingly, this study introduces sentiment analysis, a useful big data analysis tool, to understand the relevant information of online news articles and formulate an oil price trend prediction method with sentiment. Three main steps are included in the proposed method, i.e., sentiment analysis, relationship investigation and trend prediction. In sentiment analysis, the sentiment (or tone) is extracted based on a dictionary-based approach to capture the relevant online information concerning oil markets and the driving factors. In relationship investigation, the Granger causality analysis is conducted to explore whether and how the sentiment impacts oil price. In trend prediction, the sentiment is used as an important independent variable, and some popular forecasting models, e.g., logistic regression, support vector machine, decision tree and back propagation neural network, are performed. With crude oil futures prices of the West Texas Intermediate (WTI) and news articles of the Thomson Reuters as studying samples, the empirical results statistically support the powerful predictive power of sentiment for oil price trends and hence the effectiveness of the proposed method.

This paper investigates the impact of emergency order in a price-dependent newsvendor setting. To this end, we compare two ways handling the excess demand: the excess demand is lost and a penalty cost is incurred, or the excess demand can be satisfied by an emergency order. Which way is better depends on the emergency purchase cost $e$ in emergency-order way and the price $p$ plus penalty cost $g$ in lost-sales way. For a risk-neutral newsvendor, our results indicate that, when $e$ is not larger than $p+g$, the emergency order way can lead to smaller order quantity and higher expected profit. We continue to discuss the impact of newsvendor’s risk aversion and demand uncertainty on the optimal decisions of the two ways. Theoretical analysis and numerical examples indicate that when the emergency purchase cost is not high, the differentials of the optimal order quantities and expected profits will be larger as the degree of risk aversion/demand uncertainty increases. What is more, we prove that there exists a threshold value of the emergency purchase cost so that the two ways handling excess demand can obtain the same expected profit, and this threshold value increases as the degree of risk aversion decreases.

In this paper, we address the scheduling issues in a class of maintenance, repair and overhaul systems. By considering all key characteristics such as disassembly, material recovery uncertainty, material matching requirements, stochastic routings and variable processing times, the scheduling problem is formulated into a simulation optimization problem. To solve this difficult problem, we developed two hybrid algorithms based on nested partitions method and optimal computing budged allocation technology. Asymptotic convergence of these two algorithms is proved and numerical results show that the proposed algorithms can generate high quality solutions which outperform the performance of many heuristics.

In this paper, we use the meta-frontier network DEA approach to evaluate the innovation efficiency of 30 provinces in China from 2009 to 2011. These provinces have been classified into two groups based on their levels of economic development. The first group comprises provinces in the Eastern region, while the second group comprises provinces in the Central and Western regions. First, we use the meta-frontier network DEA method to estimate the technology gaps of innovation efficiency between different operating types. Second, the quadrant analysis method explores the reasons for efficiency losses. Finally, we take the fixed effect model to examine whether industry–university–research cooperation influences technology efficiency. The empirical results indicate (i) the Eastern region has significantly higher innovation efficiency than the Central and Western regions. (ii) Some Eastern provinces have a high technology level, yet their resource allocation capabilities still need to be improved. (iii) Industry–university–research cooperation is an effective way to improve innovation performance.

In this paper, we consider an online order scheduling problem with the same order size on two identical machines. The objective is to minimize the makespan. An order list $O=(O_1,O_2,\dots ,O_b)$ is given, where $b$ is a positive integer. For the problem under study, we assume that $b\ge B$ ($B=2,3,4,\dots$), that is to say, at least $B$ orders arrive. A lower bound $\frac{B+2}{B+1}$ for $b\ge B$ is obtained. Also, we design two optimal algorithms $A_1$ and $A_2$ for $B=2m$ and $B=2m+1$ ($m=1,2,3,\dots$), respectively.

Customized bundling is a pricing strategy that allows consumers to choose a certain quantity of products at a fixed price. In the reality, a customer usually has a specific rank on information goods based on their valuations, or information goods can be ranked into a list of products with decreasing valuations for a customer. Thus, we characterize customers in two dimensions for constructing the customized bundles of ranked information goods: (i) the valuation that a customer sets for his/her most favorite information good; and (ii) the total quantity of information goods with positive valuations that a customer requires. We derive the optimal customized bundling strategies in two typical scenarios and examine the impact of customer heterogeneity in terms of each dimension on the optimal pricing schemes of customized bundles. Analytical results indicate that the two features have similar effects on optimal bundle price, market penetration, and maximal profit, but impact differently on optimal bundle size. Larger customer heterogeneity leads to a lower or identical optimal bundle price, market penetration, and maximal profit. However, optimal bundle size shrinks or remains unchanged with increased customer heterogeneity on the total quantities of information goods with positive valuations, but it grows or stays the same when customers have larger heterogeneity on the valuations of their most favorite information goods. Our results provide explanations to the marketing practices of digital product firms, and also support the optimal decision of customized bundling of information goods for heterogeneous customers.

In this paper, location–allocation problem of a three-stage supply chain network, including suppliers, plants, distribution centers (DCs) and customers is investigated. With respect to the total cost, the aim is determining opened plants and DCs and designing transportation trees between the facilities. Considering the capacity of suppliers, plants and DCs are limited and there is a limitation on the maximum number of opened plants and DCs, a mixed-integer linear programming (MILP) model of the problem is presented. Since multi-stage supply chain networks have been recognized as NP-hard problems, applying priority-based encoding and a four-step backward decoding procedure, a meta-heuristic algorithm, namely GAIWO, based on the best features of genetic algorithm (GA) and invasive weed optimization (IWO) is designed to solve the problem. In small size problems, the efficiency of the GAIWO is checked by solutions of GAMS software. For larger size problems, the performance of the proposed approach is compared with four evolutionary algorithms in both aspects of the structure of the GAIWO and the efficiency of the proposed encoding–decoding procedure. Besides usual evaluation criteria, Wilcoxon test and a chess rating system are used for evaluating and ranking the algorithms. The results show higher efficiency of the proposed approach.

In this paper, we address a supply chain scheduling model with outsourcing and transportation. A job can be scheduled either on a single machine at a manufacturer’s plant or outsourced to a subcontractor. We assume that there are a sufficient number of vehicles at the manufacturer and the subcontractor such that each completed job can be transported to its customer immediately. For a given set of jobs, the decisions we need to make include the selection of jobs to be outsourced and the schedule of all the jobs. When the objective functions are to minimize the weighted sum of common scheduling measures and the total cost, we present their complexity analysis and a $2$-approximation algorithm for the second problem.

We study single-machine scheduling problems with job rejection and a deteriorating maintenance activity, where the impact of performing this activity is reflected in a reduction of the job processing times. The duration of the maintenance activity is a linear increasing function of its starting time. The aim is to determine the location of the maintenance activity and the job sequence of the accepted jobs so as to minimize scheduling cost of the accepted jobs plus total penalty of the rejected jobs. When the scheduling measures are the makespan, total completion time and combination of earliness, tardiness and due date cost, we provide polynomial time algorithms to solve these problems, respectively. When the scheduling measures are the maximum tardiness and total weighted completion time under the agreeable ratio assumption, we introduce pseudo-polynomial time algorithms to solve these ${𝒩}{𝒫}$-hard problems, respectively.