Narrow Results

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.

Nowadays, the online-to-offline (O2O) retailers provide on-demand delivery service for online orders by their own fleets and riders. An intelligent delivery network lays an important foundation to support cost-effective delivery service in the long run. Drones have great potential to revolutionize the instant delivery industry regarding cost and timeliness, while the hazard risks to humans and the environment should be seriously considered through sophisticated network design. In this paper, we propose a framework for a drone-driven intelligent delivery network design problem with the consideration of the multi-dimensional risk map, which needs to determine store location, drone fleet size and allocation, customer assignment, customer delivery mode selection, and delivery routing. A bi-objective non-linear programming model is formulated to maximize profit and minimize integrated risks as well. To tackle large instances, a modified NSGA-III algorithm is developed, which is incorporated with problem-specific search operators and Pareto local search to obtain Pareto solutions efficiently. Real-world data-based numerical experiments are conducted to verify the performance of the modified NSGA-III algorithm compared to the modified NSGA-II. A case study based on the geographical information in Shanghai is analyzed to validate the effectiveness of the proposed model. Moreover, sensitivity analysis is presented to evaluate the effects of multiple parameters on the drone delivery service network design. Some managerial insights are obtained for the O2O retailer who offers on-demand delivery service through online platform.

Credit risk evaluation has been vital for financial institutions to identify default customers and to avoid financial loss. Machine learning and data mining techniques have been adopted to develop scoring models for enhancing the prediction performance of default customers. However, it is difficult for these machine learning models for explaining the rejection or approval decision-making process to customers and other non-technical personnel. This paper presents an evidence reasoning (ER) rule-based ensemble learning approach for credit risk evaluation considering customer heterogeneity. Firstly, customers are segmented into different groups by $k$-means clustering algorithms and a two-stage weighting method is proposed to determine the significances of attributes by their discriminating powers between groups and within groups. Then, the attribute-related evidence is obtained by Bayesian statistics to represent the relationships between the attributes and credit risks, and a two-stage weighting evidential reasoning (TER) is developed as a base learner for credit scoring. Lastly, multiple base learners TERs are aggregated for evaluating customers’ credit risks. An empirical study on three credit datasets demonstrated that the proposed approach can achieve high performance with good explainability. The predicted results of the model can be well comprehended by providing the contribution of attributes and the activated rules in evidential reasoning processes.