Z. Max Shen
Prof. Z. Max Shen
Innovation and Information Management
Vice-President and Pro-Vice-Chancellor (Research)
Chair Professor in Logistics and Supply Chain Management

KB 1040

Intertemporal Pricing via Nonparametric Estimation: Integrating Reference Effects and Consumer Heterogeneity

Problem definition: We consider intertemporal pricing in the presence of reference effects and consumer heterogeneity. Our research question encompasses how to estimate heterogeneous consumer reference effects from data and how to efficiently compute the optimal pricing policy. Academic/practical relevance: Understanding reference effects is essential for designing pricing policies in modern retailing. Our work contributes to this area by incorporating consumer heterogeneity under arbitrary distributions. Methodology: We propose a mixed logit demand model that allows arbitrary joint distributions of valuations, responsiveness to prices, and responsiveness to reference prices among consumers. We use a nonparametric estimation method to learn consumer heterogeneity from transaction data. Further, we formulate the pricing optimization as an infinite horizon dynamic programming problem and solve it by applying a modified policy iteration algorithm. Results: Moreover, we investigate the structure of optimal pricing policies and prove the suboptimality of constant pricing policies even when all consumers are loss-averse according to the classical definition. Our numerical studies show that our estimation and optimization framework improves the expected revenue of retailers via accounting for heterogeneity. We validate our model using real data from JD.com, a large E-commerce retailer, and find empirical evidence of consumer heterogeneity. Managerial implications: In practice, ignoring consumer heterogeneity may lead to a significant loss of revenue. Furthermore, heterogeneous reference effect offers a strong motive for promotions and price fluctuations.

Vehicle Rebalancing in a Shared Micromobility System with Rider Crowdsourcing

Problem definition: Shared micromobility vehicles provide an eco-friendly form of short-distance travel within an urban area. Because customers pick up and drop off vehicles in any service region at any time, such convenience often leads to a severe imbalance between vehicle supply and demand in different service regions. To overcome this, a micromobility operator can crowdsource individual riders with reward incentives in addition to engaging a third-party logistics provider (3PL) to relocate the vehicles. Methodology/results: We construct a time-space network with multiple service regions and formulate a two-stage stochastic mixed-integer program considering uncertain customer demands. In the first stage, the operator decides the initial vehicle allocation for the regions, whereas in the second stage, the operator determines subsequent vehicle relocation across the regions over an operational horizon. We develop an efficient solution approach that incorporates scenario-based and time-based decomposition techniques. Our approach outperforms a commercial solver in solution quality and computational time for solving large-scale problem instances based on real data. Managerial implications: The budgets for acquiring vehicles and for rider crowdsourcing significantly impact the vehicle initial allocation and subsequent relocation. Introducing rider crowdsourcing in addition to the 3PL can significantly increase profit, reduce demand loss, and improve the vehicle utilization rate of the system without affecting any existing commitment with the 3PL. The 3PL is more efficient for mass relocation than rider crowdsourcing, whereas the latter is more efficient in handling sporadic relocation needs. To serve a region, the 3PL often relocates vehicles in batches from faraway, low-demand regions around peak hours of a day, whereas rider crowdsourcing relocates a few vehicles each time from neighboring regions throughout the day. Furthermore, rider crowdsourcing relocates more vehicles under a unimodal customer arrival pattern than a bimodal pattern, whereas the reverse holds for the 3PL.

Strengthening supply chain resilience during COVID-19: A case study of JD.com

The coronavirus/SARS-CoV-2 (COVID-19) outbreak has caused severe supply chain disruptions in practically all industries worldwide. Online e-commerce platforms, which interact directly with various industries and service numerous consumers, have become remarkable interfaces to observe the impacts of the pandemic on supply chains. Using quantitative operational data obtained from JD.com https://www.jd.com., this study analyzes the impact of the pandemic on supply chain resilience, summarizes the challenging scenarios that retailing supply chains experienced in China, and presents the practical response of JD.com throughout the pandemic. To summarize, the pandemic caused exceptional demand and severe logistical disruptions in China, and JD.com has handled well its supply chain management in response based on its integrated supply chain structure and comprehensive intelligent platforms. In particular, the existing intelligent platforms and the delivery procedures were modified slightly but promptly to deal with specific disruptions. Moreover, the entire market scenario in China was effectively controlled through the joint efforts of multiple firms, the government, and the entire Chinese society. Our study provides an example of using practical operational indicators to analyze supply chain resilience, and suggests firms pay attention to operational flexibility and collaboration beyond supply chains to deal with a large-scale supply chain disruption, such as the COVID-19 outbreak.

A Practical End-to-End Inventory Management Model with Deep Learning

We investigate a data-driven multiperiod inventory replenishment problem with uncertain demand and vendor lead time (VLT) with accessibility to a large quantity of historical data. Different from the traditional two-step predict-then-optimize (PTO) solution framework, we propose a one-step end-to-end (E2E) framework that uses deep learning models to output the suggested replenishment amount directly from input features without any intermediate step. The E2E model is trained to capture the behavior of the optimal dynamic programming solution under historical observations without any prior assumptions on the distributions of the demand and the VLT. By conducting a series of thorough numerical experiments using real data from one of the leading e-commerce companies, we demonstrate the advantages of the proposed E2E model over conventional PTO frameworks. We also conduct a field experiment with JD.com, and the results show that our new algorithm reduces holding cost, stockout cost, total inventory cost, and turnover rate substantially compared with JD’s current practice. For the supply chain management industry, our E2E model shortens the decision process and provides an automatic inventory management solution with the possibility to generalize and scale. The concept of E2E, which uses the input information directly for the ultimate goal, can also be useful in practice for other supply chain management circumstances.

New product life cycle curve modeling and forecasting with product attributes and promotion: A Bayesian functional approach

New products are highly valued by manufacturers and retailers due to their vital role in revenue generation. Product life cycle (PLC) curves often vary by their shapes and are complicated by promotional activities that induce spiky and irregular behaviors. We collaborate with JD.com to develop a flexible PLC curve forecasting framework based on Bayesian functional regression that accounts for useful covariate information, including product attributes and promotion. The functional model treats PLC curves as target variables and includes both scalar and functional predictors, capturing time-varying promotional activities. Harnessing the power of basis function transformation, the developed model can effectively characterize the local features and temporal evolution of sales curves. Our Bayesian framework can generate initial curve forecasts before the product launch and update the forecasts dynamically as new sales data are collected. We validate the superior performance of our method through extensive numerical experiments using three real-world data sets. Our forecasting framework reduces the forecasting error by 5.35%–30.76% over JD.com's current model and outperforms alternative models significantly. Furthermore, the estimated promotion effect function provides useful insights into how promotional activities interact with sales curves.

Smart Urban Transport and Logistics: A business Analytics Perspective

New technologies and innovative business models are leading to connected, shared, autonomous, and electric solutions for the tomorrow of urban transport and logistics (UTL). The efficiency and sustainability of these solutions are greatly empowered by the capability of understanding and utilizing the tremendous amount of data generated by passengers, drivers, and vehicles. In this study, we first review the innovative applications in UTL and several related research areas in the operations management (OM)/operations research (OR) literature. We then highlight the sources, types, and uses of data in different applications. We further elaborate on business analytics techniques and software developed to facilitate the planning and management of UTL systems. Finally, we conclude the paper by reflecting on the emerging trends and potential research directions in data-driven decision making for smart UTL.

Offline-Channel Planning in Smart Omnichannel Retailing

Problem definition: Observing the retail industry inevitably evolving into omnichannel, we study an offline-channel planning problem that helps an omnichannel retailer make store location and location-dependent assortment decisions in its offline channel to maximize profit across both online and offline channels, given that customers’ purchase decisions depend on not only their preferences across products but also, their valuation discrepancies across channels, as well as the hassle costs incurred. Academic/practical relevance: The proposed model and the solution approach extend the literature on retail-channel management, omnichannel assortment planning, and the broader field of smart retailing/cities. Methodology: We derive parameterized models to capture customers’ channel choice and product choice behaviors and customize a corresponding parameter estimation approach employing the expectation-maximization method. To solve the proposed optimization model, we develop a tractable mixed integer second-order conic programming reformulation and explore the structural properties of the reformulation to derive strengthening cuts in closed form. Results: We numerically validate the efficacy of the proposed solution approach and demonstrate the parameter estimation approach. We further draw managerial insights from the numerical studies using real data sets. Managerial implications: We verify that omnichannel retailers should provide location-dependent offline assortments. In addition, our benchmark studies reveal the necessity and significance of jointly determining offline store locations and assortments, as well as of incorporating the online channel while making offline-channel planning decisions.

Urban Bike Lane Planning with Bike Trajectories: Models, Algorithms, and a Real-World Case Study

Problem definition: We study an urban bike lane planning problem based on the fine-grained bike trajectory data, which are made available by smart city infrastructure, such as bike-sharing systems. The key decision is where to build bike lanes in the existing road network. Academic/practical relevance: As bike-sharing systems become widespread in the metropolitan areas over the world, bike lanes are being planned and constructed by many municipal governments to promote cycling and protect cyclists. Traditional bike lane planning approaches often rely on surveys and heuristics. We develop a general and novel optimization framework to guide the bike lane planning from bike trajectories. Methodology: We formalize the bike lane planning problem in view of the cyclists’ utility functions and derive an integer optimization model to maximize the utility. To capture cyclists’ route choices, we develop a bilevel program based on the Multinomial Logit model. Results: We derive structural properties about the base model and prove that the Lagrangian dual of the bike lane planning model is polynomial-time solvable. Furthermore, we reformulate the route-choice-based planning model as a mixed-integer linear program using a linear approximation scheme. We develop tractable formulations and efficient algorithms to solve the large-scale optimization problem. Managerial implications: Via a real-world case study with a city government, we demonstrate the efficiency of the proposed algorithms and quantify the trade-off between the coverage of bike trips and continuity of bike lanes. We show how the network topology evolves according to the utility functions and highlight the importance of understanding cyclists’ route choices. The proposed framework drives the data-driven urban-planning scheme in smart city operations management.

Real-Time Delivery Time Forecasting and Promising in Online Retailing: When Will Your Package Arrive?

Problem definition: Providing fast and reliable delivery services is key to running a successful online retail business. To achieve a better delivery time guarantee policy, we study how to estimate and promise delivery time for new customer orders in real time. Academic/practical relevance: Delivery time promising is critical to managing customer expectations and improving customer satisfaction. Simply overpromising or underpromising is undesirable because of the negative impacts on short-/long-term sales. To the best of our knowledge, we are the first to develop a data-driven framework to predict the distribution of order delivery time and set promised delivery time to customers in a cost-effective way. Methodology: We apply and extend tree-based models to generate distributional forecasts by exploiting the complicated relationship between delivery time and relevant operational predictors. To account for the cost-sensitive decision-making problem structure, we develop a new split rule for quantile regression forests that incorporates an asymmetric loss function in split point selection. We further propose a cost-sensitive decision rule to decide the promised delivery day from the predicted distribution. Results: Our decision rule is proven to be optimal given certain cost structures. Tested on a real-world data set shared from JD.com, our proposed machine learning–based models deliver superior forecasting performance. In addition, we demonstrate that our framework has the potential to provide better promised delivery time in terms of sales, cost, and accuracy as compared with the conventional promised time set by JD.com. Specifically, our simulation results indicate that the proposed delivery time promise policy can improve the sales volume by 6.1% over the current policy. Managerial implications: Through a more accurate estimation of the delivery time distribution, online retailers can strategically set the promised time to maximize customer satisfaction and boost sales. Our data-driven framework reveals the importance of modeling fulfillment operations in delivery time forecasting and integrating the decision-making problem structure with the forecasting model.