Multi-leader-follower games of freight service pricing
Open Access
- Author:
- Ansaripour, Afrooz
- Graduate Program:
- Industrial Engineering
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- February 15, 2018
- Committee Members:
- Terry L. Friesz, Dissertation Advisor/Co-Advisor
Tao Yao, Committee Chair/Co-Chair
Uday V. Shanbhag, Committee Member
Terry L. Friesz, Committee Member
Mort D. Webster, Outside Member - Keywords:
- Freight service pricing
Oligopoly
Stackelberg game
bi-level optimization
revenue management
equilibrium problem with equilibrium constraints - Abstract:
- Recently, Use of Stackelberg game and game theory concepts to solve dynamic competitive pricing, Nonlinear pricing, supply chain management and transportation network has been increased. This dissertation also focuses on Stackelberg game and EPEC formulation to deal with a linear and nonlinear pricing problem in freight service companies. We are interested in computing freight service prices when an oligopoly of freight service providers compete with one another to carry cargo for large and complex, dynamic network markets. This dissertation presents two projects as linear and nonlinear pricing on multileader- follower games of freights Service companies based on the Stackelberg-Cournot-Nash behavioral assumption. We also consider oligopoly of producers of a single abstract homogeneous commodity that is brought to market by the aforementioned freight service providers. For such an environment, we study dynamic freight service pricing from the point of view of multi-leader-follower games. The focus of the first project is on the linear pricing decision model for an oligopoly of carrier as leaders and an oligopoly of shippers as followers who compete in product’s price, production quantity output and shipments pattern. This problem is a bi-level game problem with carriers at the upper level and shippers at the lower level. To formulate the problem we review the dynamic Stackelberg games and how the levels of such games can be described as a differential variational inequalities (DVI). We also show how the DVI of the lower level can be rewritten as the mathematical complementarity formulation. Hence we are able to convert the bi-level optimization problem into a single-level problem. An application of this differential Stackelberg game will be shown in revenue management of freight services after proper time-discretization, The second topic aim to apply a more realistic, applicable and complicated model by adding non-linear pricing decision to the problem. For each carrier, the combined pricing-routing problem is a mathematical program with equilibrium constraints (MPEC). On top of this, we aim to find a Nash equilibrium among the leaders, thereby coupling multiple MPECs into a single equilibrium problem with equilibrium constraints (EPEC). We show the Computability of this EPEC model by proposing novel yet practical algorithms called double adjoint approach based on computational intelligence and high performance computing.