SCRO / Journées de l'optimisation

HEC Montréal, 29-31 mai 2023


HEC Montréal, 29 — 31 mai 2023

Horaire Auteurs Mon horaire

PAOIII Pricing and Assortment Optimization III

31 mai 2023 10h30 – 12h10

Salle: Cogeco (bleu)

Présidée par Mohammad Moshtagh

4 présentations

  • 10h30 - 10h55

    Optimal pricing strategies for new and remanufactured products with strategic customers

    • Zhuojun Liu, prés., Cape Breton University
    • Jing Chen, Dalhousie University
    • Claver Diallo, Dalhousie University
    • Michael Zhang, Saint Mary's University

    It is common for sellers to markdown product prices late in a selling season which prompts consumers to strategically manage their purchase timing. Facing consumers' strategic behaviour, the manufacturer should decide the prices for products carefully to ensure its profit. This research explores the manufacturer's optimal pricing decisions in a closed-loop supply chain by considering the customers' type, the limited number of returns, consumers' acceptability of the remanufactured products, as well as the discounting rate of consumers' perceived value for a delayed purchase. The results show that the manufacturer's retailing plan and the optimal prices depend on the unit production cost of the new and remanufactured products. In general, the consumer's strategic behaviour hurts the manufacturer's profit. When consumers are strategic, the manufacturer reduces the selling price of the new product, and it is less likely to engage in remanufacturing when the unit production cost of the new product is high as compared to the case of myopic customers. The numerical experiments provide additional implications and show that the manufacturer can take some simple counter-measure strategies to react to either the underestimation or overestimation of the expected price by the consumer.
    Keywords: Strategic customers, Pricing strategy, Remanufacturing

  • 10h55 - 11h20

    Price modeling and analysis in reverse auctions for truckload transportation services

    • Michael Haughton, prés.,
    • borzou rostami,
    • alireza amini,

    In markets for truckload transportation services, the reverse auction model involves vendors (i.e., freight transportation service providers – TSPs) operating under uncertainty when they submit bids for the prices they wish to receive for their services. This study analyzes how humans in freight transportation markets decide what price to bid for delivering a shipment, under conditions of uncertainty about factors such as the real shipment delivery cost and the bid prices being submitted by competing TSPs. The analysis addresses (a) how well the prices of human vendors align with prices suggested by rationality-based pricing models and tactics and (b) how the degree of alignment impacts vendors’ profits. The data, which were collected through extensive behavioural experiments, offer interesting insights on how pricing decisions are influenced by market factors and TSPs’ risk attitudes. Three of those insights are (i) although humans have the constraints of bounded rationality, their prices still reflect rational connections to the market factors, (ii) human pricing can be mathematically modeled as variants of rationality-based models, and (iii) the findings provide some explanation of observed pricing patterns in North American spot markets for truckload delivery services.
    Keywords: Spot market; truckload transportation; optimization; pricing decisions; behavior experiments.

  • 11h20 - 11h45

    Robust Framework for the Joint Learning of Consumer Preferences and Market Segmentation

    • Shobeir Amirnequiee, prés., Ivey Business School
    • Joe Naoum-Sawaya, Ivey Business School, Western University
    • Hubert Pun, Ivey Business School

    Assortment optimization is a core marketing task that is essential for maximizing profits. To optimize the assortment decisions, accurately segmenting the market and learning consumer preferences in each of the segments are critically important. We present a robust framework to simultaneously segment the customer base and learn each segment's preferences. We build upon ideas from machine learning and mathematical programming, and propose a robust preference elicitation model. Our model guarantees robustness against feature noise (i.e., perturbations caused by consumer misconceptions), and handles label noise (i.e., response errors) using a weighting scheme that determines the relevance of the past choices in predicting future ones. The proposed framework has three appealing characteristics. First, it simultaneously segments the market and learns the segments' preferences. Second, it extends a ML-based preference learning method that is proven to be effective. Third, the decision maker can choose the level of robustness, and has the option to focus on the parsimony of the solution. We perform extensive experiments and show that the proposed framework offers better prediction accuracy and lower variability in the predictions.

  • 11h45 - 12h10

    Optimal Markdown Policies for Perishable Products with Fixed Shelf Life

    • Mohammad Moshtagh, prés., DeGroote School of Business, McMaster University
    • Yun Zhou, DeGroote School of Business, McMaster University
    • Manish Verma, DeGroote School of Business, McMaster University

    The market for perishable products is subject to short selling seasons and volatile demand. In response, retailers often use markdowns to maximize revenue and minimize waste. Among the markdown options, the best policy is not always clear-cut, as there is a trade-off between the complexity of the policy and the revenue generated. Despite its importance, the value of different markdown policies remains understudied in the literature. To address this gap, we present a joint inventory-pricing model for perishable items with fixed shelf lives. Our study is the first to examine the effectiveness of different markdown policies, including single-stage, multiple-stage, and dynamic markdown policies, theoretically and numerically. Empirical evidence from a case study based on real-life data is used to show the performance of the proposed models. We prove that the value of markdown policies asymptotically vanishes as the shelf life, market demand, and customers' maximum willingness-to-pay increase. Conversely, the benefits of markdown policies increase when per unit expiration, shortage, and purchasing costs rise. Further, Our results suggest that while single-stage markdown policies can significantly benefit the system, in some cases, the benefit of multiple-stage markdown policies over the single-stage policies is insignificant and may not justify their complexities.