Hasso-Plattner-Institut
Prof. Dr. h.c. mult. Hasso Plattner
 

Dynamic Pricing under Competition

Modern e-commerce platforms pose both opportunities as well as hurdles for merchants. While merchants can observe markets at any point in time and automatically reprice their products, they also have to compete simultaneously with dozens of competitors.

Project Part I - Dynamic Pricing in Theory

Currently, retailers lack the possibility to test, develop, and evaluate their algorithms appropriately before releasing them into the real world. At the same time, it is challenging for researchers to investigate how pricing strategies interact with each other under heavy competition.

We built an open platform to simulate dynamic pricing competition allowing both practitioners and researchers to study the effects of automated repricing mechanisms competing with each other using market scenarios that mimic real-world marketplaces.

Interactive Simulation Platform

We built the platform in a way that one can participate and deploy own merchants with only a few lines of Python code. It allows merchants to deploy the full width of pricing strategies, from simple rule-based strategies to more sophisticated data-driven strategies using machine learning. For practitioners, the platform further provides a possibility to evaluate their pricing strategies appropriately before releasing them in production. 

To be both flexible and scalable, the platform has a scalable microservice-based architecture and is able to handle dozens of concurrent merchants and processing thousands of consumer requests per seconds. 

The platform’s source code and the technical documentation are publicly available on GitHub.

Results

Our platform enables analyses of how a strategy's performance is affected by customer behavior, price adjustment frequencies, the competitors' strategies, and the exit/entry of competitors.We compared traditional rule-based strategies with simple data-driven strategies. We find that data-driven merchants are superior to rule-based approaches as soon as a sufficiently large data set has been gathered.

 

Contact: Dr. Rainer SchlosserMartin Boissier

 

Project Part II: Dynamic Pricing in Practice

Dynamic pricing competition on modern market platforms such as Amazon Marketplace or eBay is constantly growing. Effective automated repricing systems become increasingly important for practitioners. This, however, is a highly challenging task, because offers are multi-dimensional, demand information is limited, and competitors’ price reactions are not known.

The Amazon Challenge

In this project, we partner with a top 10 seller for used books on Amazon in Germany. Our partner has an inventory of over 100,000 distinct books (ISBN), each with multiple items. The challenge is to extract as much profit as possible from a given number of books (inventory level) in a reasonable amount of time.

The pricing strategy of our project partner is characterized by a rule-based system that has been developed over the past years by carefully adjusting rules to lessons learned from selling books on Amazon. Our goal was to develop an improved data-driven pricing strategy that is based on a theoretical model.

Data-Driven Repricing Model

To be able to set up a dynamic model in order to compute optimized prices, we estimated sales probabilities. We used a logistic regression approach to quantify how offer prices and specific market situations affect sales. We consider up to 10 offer dimensions (e.g., price, quality, ratings, feedback count, shipping time) per competitor for a particular market situation.

 

The data set that we used for the regression analysis contains both the requested market situations from Amazon as well as our seller's own data (offers, sales, and inventory). Our partner requests market situations for each offered book roughly every two hours (i.e., >20 M market situations per month which result in >140 M single competitor observations per month). We used thirty customized features, e.g., the price rank of our offer price within the competitors' prices. As a result, we were able to predict sales probabilities for any offer price and for any market situation. Using efficient solution techniques, we are able to compute optimized prices for current market situations.

 

The application of our dynamic pricing strategy works as follows: First, we observe current market situations for our products, then we calculate optimized prices according to the model, and finally adjust prices on the market platform. This procedure is repeated every two hours or in case of changing market situations. This way, our strategy is able to respond immediately to new situations as prices can be adjusted in milliseconds. Moreover, new incoming sales observations are used to further improve the strategy by estimating demand more accurately.

Results

Sales results over several weeks show that our strategy outperforms the established rule-based strategy of our experienced partner regarding profitability and speed of sales, resulting in a profit increase of more than 20%.

 

Contact: Dr. Rainer Schlosser, Martin Boissier

Related Publications

  • 1.
    Groeneveld, J., Herrmann, J., Mollenhauer, N., Dreessen, L., Bessin, N., Schulze-Tast, J., Kastius, A., Huegle, J., Schlosser, R.: Self-Learning Agents for Recommerce Markets. Business & Information Systems Engineering 66 (4). 441–463 (2024).
     
  • 2.
    Schlosser, R., Gönsch, J.: Risk-Averse Dynamic Pricing using Mean-Semivariance Optimization. European Journal of Operational Research. 310 (1), 1151–1163 (2023).
     
  • 3.
    Schlosser, R., Chenavaz, R.: Joint Dynamic Pricing and Marketing-Mix Strategies for Revenue Management Applications with Stochastic Demand. International Transactions in Operational Research, accepted. (2023).
     
  • 4.
    Kastius, A., Schlosser, R.: Towards Transfer Learning for Revenue and Pricing Management. Operations Research Proceedings, OR2021. bll. 361–366 (2022).
     
  • 5.
    Kastius, A., Schlosser, R.: Dynamic Pricing under Competition using Reinforcement Learning. Journal of Revenue and Pricing Management. 21, 50–63 (2022).
     
  • 6.
    Schlosser, R.: Heuristic Mean Variance Optimization in Markov Decision Processes using State-Dependent Risk Aversion. IMA Journal of Management Mathematics. 33 (2), 181–199 (2022).
     
  • 7.
    Chenavaz, R., Klibi, W., Schlosser, R.: Dynamic Pricing with Reference Price Effects in Integrated Online and Offline Retailing. International Journal of Production Research. 60, 5854–5875 (2022).
     
  • 8.
    Schlosser, R., Kastius, A.: Stochastic Dynamic Pricing under Duopoly Competition with Mutual Strategy Adjustments. Operations Research Proceedings (OR 2021). bll. 367–372 (2022).
     
  • 9.
    Schlosser, R., Westphal, J., Pörschke, M., Maltenberger, T., Kaminsky, Y.: Self-Adaptive Agents in a Dynamic Pricing Duopoly: Competition, Collusion, and Risk Considerations. Springer Nature Computer Science. 3 (3), 1–17 (2022).
     
  • 10.
    Figge, F., Dimitrov, S., Schlosser, R., Chenavaz, R.: Does the circular economy fuel the throwaway society? The role of opportunity costs for products that lose value over time. Journal of Cleaner Production. 368 (133207), (2022).
     
  • 11.
    Schlosser, R., Kastius, A.: A Conceptual Framework for Studying Self-Learning Agents in Recommerce Markets. Operations Research Proceedings (OR 2022), to appear (2022).
     
  • 12.
    Kastius, A., Schlosser, R.: Multi-Agent Dynamic Pricing Using Reinforcement Learning and Asymmetric Information. Operations Research Proceedings (OR2022), to appear (2022).
     
  • 13.
    Schlosser, R., Chenavaz, R., Dimitrov, S.: Circular Economy: Joint Dynamic Pricing and Recycling Investments. International Journal of Production Economics. 108117, 1–13 (2021).
     
  • 14.
    Kaminsky, Y., Maltenberger, T., Pörschke, M., Westphal, J., Schlosser, R.: Pricing Competition in a Duopoly with Self-Adapting Strategies. 10th International Conference on Operations Research and Enterprise Systems (ICORES 2021). bll. 60–71 (2021).
     
  • 15.
    Schlosser, R.: Scalable Relaxation Techniques to Solve Stochastic Dynamic Multi-Product Pricing Problems with Substitution Effects. Journal of Revenue and Pricing Management. 20 (1), 54–65 (2021).
     
  • 16.
    Schlosser, R.: Risk-Sensitive Control of Markov Decision Processes: A Moment-Based Approach with Target Distributions. Computers and Operations Research. 123 (104997), 1–15 (2020).
     
  • 17.
    Schlosser, R.: Stochastic Dynamic Pricing with Waiting and Forward-Looking Consumers. Communications in Computer and Information Science (CCIS), Vol. 1162. bll. 47–69. Springer (2020).
     
  • 18.
    Schlosser, R., Richly, K.: Dynamic Pricing Competition with Unobservable Inventory Levels: A Hidden Markov Model Approach. Communications in Computer and Information Science. bll. 15–36. Springer (2019).
     
  • 19.
    Schlosser, R.: Data-Driven Stochastic Dynamic Pricing and Ordering. Operations Research Proceedings 2018. bll. 397–403 (2019).
     
  • 20.
    Schlosser, R., Walther, C., Boissier, M., Uflacker, M.: Automated Repricing and Ordering Strategies in Competitive Markets. AI Communications. 32, 15–29 (2019).
     
  • 21.
    Schlosser, R.: Stochastic Dynamic Pricing with Strategic Customers and Reference Price Effects. 8th International Conference on Operations Research and Enterprise Systems, ICORES 2019. bll. 179–188 (2019).
     
  • 22.
    Schlosser, R., Richly, K.: Dynamic Pricing under Competition with Data-Driven Price Anticipations and Endogenous Reference Price Effects. Journal of Revenue & Pricing Management. 18, 451–464 (2019).
     
  • 23.
    Schlosser, R., Boissier, M.: Dynamic Pricing under Competition on Online Marketplaces: A Data-Driven Approach. Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, KDD. bll. 705–714 (2018).
     
  • 24.
    Schlosser, R., Walther, C., Boissier, M., Uflacker, M.: Data-Driven Inventory Management and Dynamic Pricing Competition on Online Marketplaces. Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI 2018). bll. 5856–5858 (2018).
     
  • 25.
    Schlosser, R., Richly, K.: Dynamic Pricing Strategies in a Finite Horizon Duopoly with Partial Information. 7th International Conference on Operations Research and Enterprise Systems, ICORES 2018. bll. 21–30 (2018).
     
  • 26.
    Schlosser, R.: Stochastic Dynamic Multi-Product Pricing under Competition. Operations Research Proceedings 2017. bll. 527–533 (2018).
     
  • 27.
    Schlosser, R., Boissier, M.: Optimal Repricing Strategies in a Stochastic Infinite Horizon Duopoly. Communications in Computer and Information Science (CCIS). bll. 129–150. Springer (2018).
     
  • 28.
    Schlosser, R., Boissier, M.: Dealing with the Dimensionality Curse in Dynamic Pricing Competition: Using Frequent Repricing to Compensate Imperfect Market Anticipations. Computers and Operations Research. 100, 26–42 (2018).
     
  • 29.
    Schlosser, R., Boissier, M.: Optimal Price Reaction Strategies in the Presence of Active and Passive Competitors. Proceedings of the 6th International Conference on Operations Research and Enterprise Systems (ICORES), Porto, Portugal. bll. 47–56 (2017).
     
  • 30.
    Schlosser, R.: Stochastic Dynamic Pricing and Advertising in Isoelastic Oligopoly Models. European Journal of Operational Research. 259, 1144–1155 (2017).
     
  • 31.
    Boissier, M., Schlosser, R., Podlesny, N., Serth, S., Bornstein, M., Latt, J., Lindemann, J., Selke, J., Uflacker, M.: Data-Driven Repricing Strategies in Competitive Markets: An Interactive Simulation Platform. Proceedings of the Eleventh ACM Conference on Recommender Systems (RecSys ’17). bll. 355–357. ACM, New York, NY, USA (2017).
     
  • 32.
    Serth, S., Podlesny, N., Bornstein, M., Lindemann, J., Latt, J., Selke, J., Schlosser, R., Boissier, M., Uflacker, M.: An Interactive Platform to Simulate Dynamic Pricing Competition on Online Marketplaces. 21st IEEE International Enterprise Distributed Object Computing Conference, EDOC 2017, Quebec City, QC, Canada, October 10-13, 2017. bll. 61–66. IEEE (2017).
     
  • 33.
    Schlosser, R.: Stochastic Dynamic Multi-Product Pricing with Dynamic Advertising and Adoption Effects. Journal of Revenue and Pricing Management. 15, 153–169 (2016).
     
  • 34.
    Schlosser, R., Boissier, M., Schober, A., Uflacker, M.: How To Survive Dynamic Pricing Competition in E-commerce. Proceedings of the Poster Track of the 10th ACM Conference on Recommender Systems (RecSys 2016), Boston, USA, September 17, 2016 (2016).
     
  • 35.
    Schlosser, R.: Joint Stochastic Dynamic Pricing and Advertising with Time-Dependent Demand. Journal of Economic Dynamics and Control. 73, 439–452 (2016).
     
  • 36.
    Schlosser, R.: Dynamic Pricing with Time-Dependent Elasticities. Journal of Revenue and Pricing Management. 14, 365–383 (2015).
     
  • 37.
    Schlosser, R.: A Stochastic Dynamic Pricing and Advertising Model under Risk Aversion. Journal of Revenue and Pricing Management. 14, 451–468 (2015).
     
  • 38.
    Schlosser, R.: Dynamic Pricing and Advertising Models with Inventory Holding Costs. Journal of Economic Dynamics and Control. 57, 163–181 (2015).
     
  • 39.
    Helmes, K., Schlosser, R.: Dynamic Advertising and Pricing with Constant Demand Elasticities. Journal of Economic Dynamics and Control. 37, 2814–2832 (2013).