Summer School

New Paradigms in Revenue Management: Methods and Applications

May 25 – 27, 2016

University of Maryland

On this page...

Topics
Organizers
Lecturers
Location
Contact
Program
Registration

This Summer School Program will provide basic and advanced methodologies for revenue management problems related to the transportation sector. The course aims at integrating different disciplines to solve both theoretical and practical concerns that arise in the design and development of systems for network revenue management, pricing, and operation. The lectures will cover operation research methods, optimization algorithms and econometric models.  Particular emphasis will be given to the implementation of behavioral models, based on the theory of discrete choice models (random utility maximization) into revenue management. Case studies related to revenue management problems will illustrate how these methods can be applied in practice. Methods that have been traditionally applied to the airline sector will be transferred to other types of carriers (especially rail) and their potential for generating new demand and more revenue will be assessed. Keynotes will be given on timely topics by well known researchers in the field. The summer school is open to Master, Ph.D. and post doctoral students, researcher and practitioners and will facilitate professional networking and exchange of ideas about the theory and practice of research in revenue management.

Topics

  • Network Revenue Management
  • Behavioral Models and Discrete Choice Analysis
  • Stochastic Programming
  • Dynamic Programming
  • Non-Linear Optimization
  • Decision Support Systems for revenue management
  • Real case studies

Organizers

Cinzia CirilloCinzia Cirillo
ccirillo@umd.edu

Cinzia Cirillo is Associate Professor at the Department of Civil & Environmental Engineering of the University of Maryland. Dr. Cirillo's research is in the area of transportation demand analysis, forecasting, and planning.  Her work includes the development of improved transport modeling techniques and survey instruments, application of advanced statistical and econometrics methods, and analysis of their results to predict consumer demand and behavior for various transportation options.

Shadi AzadehShadi Sharif Azadeh
shadi.sharifazadeh@epfl.ch

Shadi Sharif Azadeh holds a Postdoc position at TRANSP-OR at EPFL. She had completed her PhD in Mathematics (Operations Research). Her field of research is demand modeling, more specifically, choice-based optimization models. She is also affiliated to ITS lab at MIT, working on the application of choice modeling and revenue management in the context of Flexible Mobility on Demand (FMOD).

Top of page

Lecturers

  • Fabian Bastin (University of Montreal, CA)
  • Emmanuel Carrier (Data Science Consultant at Delta Air Lines)
  • Cinzia Cirillo (University of Maryland, USA)
  • Laurie Garrow (Georgia Tech, USA)
  • Mikhail Nediak (Queens University, CA)
  • Jonathan Owen (Director of Operations Research at General Motors)
  • Sergey Shebalov (Senior Director of Operations Research at Sabre)
  • Shadi Sharif Azadeh (EPFL, Switzerland)

Location

  • University of Maryland
    Kim Engineering Building

Contact

  • Cinzia Cirillo
    Department of Civil & Environmental Engineeringccirillo@umd.edu
    University of Maryland
    3250 Kim Engineering Building
    College Park MD  20742
    301.405.301.6864
    ccirillo@umd.edu

Top of page


PROGRAM

WEDNESDAY, May 25, 2016

9:00-9:30 WELCOME AND BREAKFAST

9:30-10:45 (Cinzia Cirillo)
Introduction of Discrete Choice Modeling for Revenue Management
Choice models based on random utility theory are being used in revenue management problems because of their ability to deal with customer heterogeneity and preferences over a set of multiple products. In this lecture, multinomial logit models, latent class models and mixture of logit are proposed to model ticket purchase timing over a finite sale horizon; in particular, parametric and non-parametric mixed logit are formulated and estimated. The analysis relative to intercity railway trips is performed on real data extracted from internet booking records, which contain very limited information that can be used for customers’ segmentation. The parameter estimates are then integrated into a nonlinear optimization framework, and fare strategies and seat allocations are derived.

10:45-11:00 COFFEE BREAK

11:00-12:15 (Fabian Bastin)
Introduction to stochastic dynamic programming
In this talk, we will introduce basic concepts of dynamic programming, focusing on the frequent situation where uncertainty affects the decision process. The uncertainty will be represented by means of random variables, and we will give some guidelines to apply stochastic dynamic programming in the context of revenue management. We will also underline some of the challenges currently faced by researchers and practitioners.

12:15-13:30 LUNCH

13:30-14:45 (Shadi Sharif)
Demand modeling in revenue management
The purpose of revenue management (RM) is to enhance the profitability of a firm through the optimal management of its inventory. In transportation industry, this can be achieved by controlling the availability of products (e.g. different fare classes) as well as setting up pricing policies in order to redirect customers (passengers) to products (fare classes) with high profit margins. Any such strategy is highly dependent on demand models.

During this presentation, I would like point out the challenges researchers face in modeling demand and how optimization techniques could be applied to tackle such problems.

14:45-15:15 COFFEE BREAK

15:15-16:30  (Laurie Garrow)
Accounting for price endogeneity in airline itinerary choice models
This study formulates an itinerary choice model that is consistent with those used by industry and corrects for price endogeneity using a control function that uses several types of instrumental variables. We estimate our models using database of more than 3 million tickets provided by the Airlines Reporting Corporation. Results based on Continental U.S. markets for May 2013 departures show that models that fail to account for price endogeneity overestimate customers’ value of time and result in biased price estimates and incorrect pricing recommendations.

Top of page


THURSDAY, MAY 26, 2016

8:30-9:00 BREAKFAST

9:00-10:30 KEYNOTE LECTURE 1
Trey Romero, AMTRAK, Director of Revenue Management.

 

10:30-11:00 COFFEE BREAK

11:00-12:30 KEYNOTE LECTURE 2
Management and development of urban toll road networks
Jennifer Aument

TRANSURBAN - Group General Manager, North America
 

12:30-13:30 LUNCH

13:30-15:00 (Mikhail Nediak, joint work with Jue Wang & Yuri Levin) Queens University, (CA)
Dynamic Choice Models of Strategic Customer Behavior for Dynamic Pricing in Continuous Time

In this talk, we present strategic behavior models where customers make their choices at the time points generated by a Poisson process and the chosen decision/action affects customer state.  Given customer expectations about the future evolution of the environment, this approach provides a straightforward model of the customer utilities in each state and the corresponding customer population dynamics in the context of dynamic pricing and other demand management policies.

15:00-15:15 COFFEE BREAK

15:15-16:30 (Jonathan Owen)
Automotive Applications of Revenue Management
This talk describes a variety of problem contexts and practical considerations for implementation of revenue management analytics at a large automotive OEM.  Examples will be provided that build on a blend of both observed and stated consumer preferences at different points in the vehicle development and sales process.

Top of page


FRIDAY, MAY 27, 2016

8:30-9:00 BREAKFAST

9:00-10:30 (Sergey Shebalov)
Offer Management

Airlines are continually reinventing their revenue management and distribution practices. One of the major recent trends is related to New Distribution Capabilities championed by IATA. It includes several fundamentally new components such as Dynamic Pricing, Customer Centricity, Airline Retailing, etc. We will discuss new requirements this initiative imposes on revenue management decision support systems, new problems Operations Research Community would need to address and several techniques and approaches that can be employed to overcome these challenges.

10:30-11:00 COFFEE BREAK

11:00-12:30 (Emmanuel Carrier)
The Practice of Revenue Management: Challenges and Opportunities.
Revenue management has been one of the most successful management science techniques and has gradually spread to a growing number of industries. While RM has brought large financial benefits to firms, it has been a win-win strategy and also provided many benefits to consumers such as lower median prices and society at large through a better utilization of capital-intensive and energy-intensive assets. However, implementing a successful RM strategy has proved challenging for many firms. In this presentation, we will discuss some of the most common pitfalls from counter-intuitive effects to the delicate balance between man and machine and how to best overcome them.


Registration

The early registration fee (until May 10th, 2016) is $200 for students and $500 for faculty/professionals. After the registration fees are respectively, $250 and $600. The fee covers coffee breaks and lunches. Please note that the number of participants is limited.

Online Registration

Cancellations must be received in writing; no refunds will be issued after May 20, 2016.

The National Transportation Center at University of Maryland sponsors the registration fee for five graduate students. In order to apply, please send your CV and a letter of motivation (maximum one page) to Cinzia Cirillo (ccirillo@umd.edu) before April 20th, 2016.

Top of page