ORIE 4154 - Revenue Optimization and Marketplace Design
Course Description
This course aims to expose students to modeling and optimization techniques which can be used to design the firm-market interface. In particular, this interface can take two forms:
- Demand management: How to sell the right product to the right customer at the right time for the right price.
- Marketplace design: Determining who gets what and why (and at what price) in two-sided platform marketplaces.
We will draw on a rich body of research spanning three disciplines - operations management, economics and computer science - and study a collection of related (yet diverse) mathematical models and techniques. The models we cover have proved successful in practice across various industries; however, given the transformations brought about by the advent of online commerce and increased use of smartphones, our aim is to provide students with tools that generalize to new domains.
Course Information
- Syllabus: pdf
- Lectures: TR 11.40pm-12.55pm, Thurston 205, Map
- Prelim: 28th March 2017, 7.30pm - 9.30pm, Kimball B11, Map
- Instructor: Siddhartha Banerjee, 229 Rhodes Hall, email
- Teaching Assistants:
Assignments
- Homework 1 (Due by 11.59pm on Friday, 10th February)
- Homework 2 (Due by 11.59pm on Friday, 28th February)
- Homework 3 (Due by 11.59pm on Wednesday, 15th March)
- Homework 4 (Due by 11.59pm on Sunday, 26th March)
- Homework 5 (Due by 11.59pm on Wednesday, 5th May)
Lectures
- Lecture 1: Introduction
- Slides: Lecture 1
- Lecture 2: Single fare-class capacity allocation (Littlewood's Law)
- Slides: Lecture 2
- Lecture 3: Intro to dynamic programming
- Slides: Lecture 3
- Lecture 4-7: Multiple fare-class capacity allocation
- Slides: Lecture 4
- Notes: Optimality of protection-levels
- Notes: Computing protection levels
- Lecture 8: Intro to network revenue management
- Slides: Lecture 8
- Lecture 9: The Network RM Dynamic Program
- Slides: Lecture 9
- Lecture 10: LP approximation for single-resource allocation problems
- Notes: LP-based approximations
- Lecture 11: Convexity and LP-based Approximations
- Notes: Randomized/fluid LP bounds
- Lecture 12: LP-duality and bid prices for network RM
- Notes: Bid-prices for Network RM
- Lecture 12.5: Snow day...
- Lecture 13: The perils of optimizing under the wrong model
- Notes: Spiral-down effect
- Lecture 14: Choice Models
- Lecture 15: Assortment Optimization under the MNL model
- Notes: Assortment Optimization
- Lecture 16,17: Constrained assortment optimization under the MNL model
- Lecture 18: Strategic customer-choice models and mechanism design
- Notes: Strategic customer models
- Lecture 19,20: Single-parameter environments and Myerson's Lemma
- Notes: Myerson's Lemma
- Lecture 21,22: Optimal revenue mechanisms
- Notes: Optimal revenue mechanisms
- Lecture 23,24: Pricing in two-sided marketplaces
- Notes: Two-sided markets
- Lecture 25: Course summary, and beyond...
- Notes: Course summary
References
There is no required textbook for the course; we will cover materials from a variety of sources, and relevant notes and references will be uploaded here and on Piazza. That said, for interested students, here are some recommended books.
- The following textbooks are a great resource for students interested in reading more about the history, organizing principles, and applications of pricing and revenue management:
- Pricing and Revenue Optimization by Robert Phillips: This gives an excellent overview of the operational viewpoint of revenue management, though from a less technical perspective than what we will cover.
(The Cornell library gives you access to an online version) - Principles of Pricing by Vohra and Krishnamurthi: This looks at pricing from an economics and marketing perspectives. Excellent for background reading, but again, at a less technical level than what we will cover.
- Pricing and Revenue Optimization by Robert Phillips: This gives an excellent overview of the operational viewpoint of revenue management, though from a less technical perspective than what we will cover.
- An excellent, though somewhat technical book for the main revenue management results which we will cover in class is:
- The Theory and Practice of Revenue Management by Kalyan Talluri and Garrett van Ryzin.
(The Cornell library gives you access to an online version.)
- The Theory and Practice of Revenue Management by Kalyan Talluri and Garrett van Ryzin.
- For the market design segment, the following two books are excellent references (and available online!).
- Twenty Lectures on Algorithmic Game Theory by Tim Roughgarden: We will cover the first 8 lectures; however this is an excellent book for game theory in general. You can also refer to Tim's original lecture notes.
- Mechanism Design and Approximation by Jason Hartline: This covers the same material, but in much greater detail (and at a somewhat more technical level).
The course will draw extensively on earlier versions of this course taught by Huseyin Topaloglu and Paat Rusmevicheintong; some material related these courses may be uploaded on Blackboard as the semester progresses.