This course aims to introduce the use of behavioral and experimental methods that have been increasingly popular in the field of operations management. In particular we use a supply chain scenario to study how human beings make individual and strategic decisions in the face of uncertainty and risk. First, we use the standard newsvendor problem to discuss decisions involving only a single individual. This problem is concerned with the order quantity decision of a retailer that faces probabilistic demand. Then, we consider a simple manufacturer-retailer supply chain where the retailer faces the newsvendor problem, and her problem parameters are determined by the contract that the manufacturer offers. This scenario allows us to study what happens to decisions when two individuals interact strategically with each other . Course discussion is based on results from decision-making experiments with human subjects. In addition to published research papers, we also use data from experiments conducted at Sabanci University.

In this course, we study how human beings make decisions in the face of uncertainty and risk. We investigate individual and strategic decision-making issues using a supply chain context. 

• First, we use the standard newsvendor problem to discuss decisions involving only a single individual. This problem is concerned with the order quantity decision of a retailer that faces probabilistic demand.
• Then, we consider a simple manufacturer-retailer supply chain where the retailer faces the newsvendor problem, and her problem parameters are determined by the “contract” (the wholesale price, etc.) that the manufacturer offers. This scenario allows us to study what happens to decisions when two individuals interact strategically with each other. 
   

To understand human behavior, we study experimental research that is based on decision-making experiments with human subjects. In addition to published research papers, we also use data from our own experiments conducted here at Sabanci University. In these experiments, student subjects play the roles of manufacturer and retailer, making contract and order quantity decisions for around 40 rounds. Comparing experimental data with theoretical model predictions, we aim to answer questions such as:

• Do subjects make decisions as predicted by theory? If not, what are the factors that affect their decisions?
• Do subjects use certain decision heuristics in making decisions?
• Can we build statistical models to better predict human behavior?
• Do different subject pools exhibit different behavior?
• Do subjects learn to make better decisions over time?
   

We also study the effects of strategic human interaction between the two parties in a supply chain. Relevant questions include

• Do the subjects behave as predicted by the game-theoretical equilibrium?
  Can the manufacturer anticipate the retailer’s reaction when determining contract parameters?
• Do manufacturers care about fairness when offering the contract?
• What factors affect the retailer’s contract rejection decision?
• Can we explain subject decisions using personality data? 
   

We consider these questions under different supply chain contracts, including wholesale price, buyback, and revenue sharing contracts. We aim to compare the experimental performance of these contracts with each other and also with theoretical predictions. 

These topics have important practical implications in the design of decision support systems, incentive systems, as well as supply chain contracts. Understanding human decision-making under risk and in strategic interactions is also a prerequisite for developing successful Artificial Intelligence platforms that will work with human managers.

 

Topics

• Introduction:
  • Decision experiments in Operations Management
  • Introduction to Supply Chain Experiments & their data
• Part I: Individual decisions
  • The newsvendor model
  • Newsvendor experiments
  • The "Pull to center" effect, decision heuristics
  • Effects of learning and feedback
  • Behavioral models to explain newsvendor decisions
  • Subject-level versus average analysis
• Part II: Strategic interaction between individuals
  • Supply chain coordination and the double marginalization problem
  • Game-theoretic analytical solutions under different contracts
  • Comparing analytical predictions with experimental data
  • Building behavioral models to explain manufacturer and retailer decisions
  • Using subjects’ personality data to explain decisions.