Packed Up and Off to the INFORMS Computing Society Conference in Richmond

Having a conference such as the 2015 INFORMS Computing Society Conference, which takes place in Richmond, Virgina, January 11-13, 2015,  is a treat.

The theme of the conference is: Operations Research and Computing: Algorithms and Software for Analytics.

The timing of the conference is convenient for faculty and students since it is taking place during the winter break between academic semesters.

I have packed up my suitcase and briefcase and will soon be heading to the airport.

The weather in Richmond is much milder than here in western Massachusetts so that will make the conference pleasant as well.

The conference organizing committee and the stream chairs did a terrific job putting the program together There will be many methodological sessions as well as application-based ones - I am especially interested in the network applications, the homeland security, and the  health applications tracks. There is also a marvelous lineup of plenary speakers and tutorial speakers.  The first plenary speaker is Professor George Nemhauser of Georgia Tech, tomorrow morning - a talk not to be missed.

The program and the Proceedings can be downloaded from this page.

On Tuesday morning, in a session in the Network Applications track, I will be presenting a paper, with assistance from my doctoral student, Sara Saberi, entitled: "A Game Theory Model for a Differentiated Service-Oriented Internet with Duration-Based Contracts." This paper is joint with also Professor Tilman Wolf of the College of Engineering at UMass Amherst and Professor Ladimer S. Nagurney from the Department of Electrical and Computer Engineering at the University of Hartford.

Our paper can be accessed from the online Proceedings, compliments of INFORMS.

Many thanks to the Editors of this volume: Drs. Brian Borchers,  J. Paul Brooks, and Laura McLay!

Our presentation can be downloaded from the Supernetwork Center site here.

Looking forward to seeing many INFORMS Computing Society members soon in Richmond, Virginia!

Offering More Choices on the Internet Through Duration-Based Contracts

We, as consumers, are locked into contracts with Internet Service Providers (ISPs)  over time durations that are too rigid,  over periods of yesrs, typically.

Wouldn't it be wonderful if consumers could be offered more choices, with choices including having contracts for Internet services for the time period that you actually want and need? Perhaps you just want to catch up on the news or view a video or movie.

This is one of the characteristics of a Future Internet Architecture (FIA) that we have been working on as part of our multiuniversity National Science Foundation (FIA) project: Network Innovation Through Choice. Specifically, we have been researching both theoretically and in terms of implementation an architecture which we are calling ChoiceNet with the goal being of introducing an Economy Plane for the Internet. The three principles of ChoiceNet are depicted graphically below.

In one of our fundamental conceptual papers, written by all the investigators on this project, we note that: These three principles interact in a cyclic process that results in increased competition, a faster pace of innovation, and better information for consumers, making it possible for users to make more informed decisions. We expect this “virtuous cycle” to be repeated over and over, on much smaller time scales than is possible in today’s Internet, where customers are locked into their local provider and rely on the latter for everything. Furthermore, we note that omitting any element of the cycle destroys the effectiveness of the others.

In order to investigate duration-based contracts for the Internet, we developed a game theory model for a differentiated service-oriented Internet with duration-based contracts. The model uses variational inequality theory for the formulation of the equilibrium in service usage rates, quality, and contract durations and projected dynamical systems theory for the underlying competitive dynamics. This paper we will be presenting at the 2015  INFORMS Computing Society Conference in Richmond,, VA in mid-January.   The paper will also be published in the conference proceedings. My co-authors on this paper are: Professor Tilman Wolf, who is the Lead PI on our NSF project, Sara Saberi, one of my doctoral students in Management Science at the Isenberg School, who this year received the prestigious 2014 Isenberg Scholar Award, and also Professor Ladimer S. Nagurney, whose university posted a writeup on our paper.

And for those of you who can make it, next Friday, Professor Tilman Wolf will be delivering a talk on ChoiceNet at the Isenberg School of Management in our great UMass Amherst INFORMS Speaker Series. I hope that you can join us!

Net Neutrality and Opening Up the Black Box

I very much care about fairness, equity, and making sure that there is no discrimination and that also is the case for the Internet.

Net neutrality is a topic that is now being fiercely debated on the news and talk shows and has many taking political sides.


To many consumers, and even businesses, the Internet is a “black box,” which has revolutionized the way in which economic and social transactions and interactions are conducted, entertainment is obtained and experienced, communications are done, and numerous educational activities participated in. If one were to ask Internet users to identify the mechanisms by which the videos that they view, the news that they read online, and the messages that they disseminate, one may very well find that a consumer would be able to identify the content provider and, perhaps, the network provider. The user/consumer will not know where the content is being delivered from, the routes that have been taken to transmit the data, and the service providers that might have been transacted with en route. Moreover, they will not be aware of the encumbered costs and, very likely, the profit garnered.

Hence, we are dealing with information asymmetry in today’s Internet, a subject that, in terms of quality, dates to the Nobel laureate’s George Akerlof’s classic 1970 paper. Consumers, as well as businesses, are unaware of the intermediaries involved in network provision, the costs and profits of service providers, and are locked into contracts of durations that are inflexible and, possibly, not optimal from their utility perspectives. Such an economic landscape creates an inertia for innovation as well as consumer satisfaction and limits competitive behavior. Furthermore, it is not clear, due to the lack of transparency and information asymmetry, whether scarce resources are being utilized in an optimal manner, with the understanding that different stakeholders and decision-makers may be faced with distinct objective functions, which may, include, for example, aspects of environmental sustainability as well as cybersecurity and even enhanced quality for network transport provision.

Furthermore, the constraints that stakeholders and decision-makers are faced with are not transparent, which may result of an inefficient use of resources. The information asymmetry results in a lack of trust, and, even to a certain extent, paranoia on the consumers’ part since vital economic and engineering information is not shared in a transparent and clear manner.


The topic inspired us to write a Guest Commentary entitled, "Net Neutrality, the Past, Present, Future and What It Means to Us"  for masslive.com


My colleagues and I have been working since 2011 on a project funded by the National Science Foundation (NSF), Network Innovation Through Choice. As part of the project, we are envisioning the components of an economy plane for Future Internet Architectures and, in particular, one which we are developing and which we are calling ChoiceNet. 

Specifically, we wish to offer consumers more choices and in a transparent way. Why should we be locked into long-term contracts, for example, when it comes to Internet services?!

Towards that end, we have proposed duration-based contracts for a service-oriented Internet. In the paper, A Game Theory Model for a Differentiated Service-Oriented Internet with Duration-Based Contracts, co-authored with my doctoral student, Sara Saberi, Professor Tilman Wolf, the PI on our NSF project, and Professor Ladimer S. Nagurney, we introduced a game theory model of a service-oriented Internet in which the network providers compete in usage service rates, quality levels, and duration-based contracts. We formulate the network-based Nash equilibrium conditions as a variational inequality problem, provide qualitative properties of existence and uniqueness, and describe an algorithm, which yields closed-form expressions at each iteration. The numerical examples include sensitivity analysis for price functions at the demand markets as well as variations in the upper bounds on the quality levels for the services.

The paper will be presented at the INFORMS Computing Society Conference in January 2015 in Richmond, VA and will be published in its Proceedings.

And speaking of quality, our paper,  A Network Economic Game Theory Model of a Service-Oriented Internet with Price and Quality Competition in Both Content and Network Provision, appears in the December 2014 issue of the INFORMS journal Service Science.

In the meantime, I am sure that the lively discussions on net neutrality and regulations will continue!

Transparency is key!