Computational Social Science and Supply Chain Networks

I had a great time giving a seminar yesterday on Grand Challenges and Opportunities in Supply Chain Networks: From Analysis to Design in the Computational Social Science Initiative Seminar Series at UMass Amherst.  The presentation took place on the 9th floor of the Campus Center in the middle of the campus and the view was fabulous on a bright, sunny, and actually warm day in February.

Yahoo is giving financial support for the seminar series and it was nice that lunch was provided by UMass catering.

Several of my doctoral students from the Isenberg School of Management came and it was great to see sociologists, computer scientists, and engineers in the audience, as well.

I focused on the importance of capturing the behaviors of the decision-makers in the context of supply chains and also overviewed the Braess paradox (classic (1968) version) and showed photos of Professor Braess visiting the Isenberg School, after our translation of his paper from German to English, along with Tina Wakolbinger, appeared in Transportation Science.  I showed how we used evolutionary variational inequalities to demonstrate how the Braess paradox only occurred on the classic Braess network for a range of demands. This was work that I did while I was  a 2005-2006 Fellow at the Radcliffe Institute for Advanced Study at Harvard University. My co-authors on this paper were Professor David Parkes of Harvard and Professor Patrizia Daniele of the University of Catania. The paper appeared in Computational Management Science. 

In addition, I had to mention our work on the integration of social networks with supply chains and with financial networks, as well as the evolution of the product flows, relationship levels, and prices over time. This was also work done with Tina Wakolbinger.


Coincidentally, Tina Wakolbinger, who is now a Full Professor (I am so proud of my former students)  is flying into Boston today to take part in the Dynamics of Disasters symposium that I organized for the AAAS meeting that is taking place in Boston, February 14-18, 2013. Our symposium will be at the Hynes Center tomorrow afternoon and I can hardly wait -- the panelists and discussants are simply superb -- Professor David McLaughlin, Professor Laura McLay, and Professor Panos M. Pardalos, with Professors Tina Wakolbinger and Jose Holquin-Veras as discussants.

The questions from the audience at my seminar were really good (Thank you!) and ranged from questions on existence and uniqueness of solutions and multiple equilibria as well as stability analysis, ongoing and future research, as well as how the methodologies that we have been instrumental in co-developing (projected dynamical systems, for example) have migrated to different disciplines, including neuroscience. I had highlighted the wide range of supply chains that we had worked on from electric power ones with empirical results for New England to humanitarian and healthcare ones for critical needs products and pharmaceutical products, respectively, and blood supply chains.  I even spoke on network synergies and assessment in the context of mergers and acquisitions. M&As have been big news lately with the American Airlines and USAir merger (still needs approval by regulators), Berkshire Hathaway and Heinz -- breaking related news on this one, and others, including Dell.

I was also asked about our work on supply chains in nature, so I had the opportunity to speak on work that we have done with Professor Christian Mullon of France on the network economics of ecological systems with data from marine ecosystems. I mentioned his forthcoming book, which he graciously forwarded a recent draft of to me. The book, which should be available this July, is called Network Economics of Marine Ecosystems and their Exploitation,  He acknowledges me in the book for "inspiration." To see network economics, variational inequalities, and projected dynamical systems being utilized in this novel application domain is simply thrilling.

I concluded my lecture by showing by latest book, Networks Against Time: Supply Chain Analytics for Perishable Products, hardcopies of which had arrived on Valentine's Day, to the audience.

Special thanks to Professors Ryan Acton and James Kitts for the great hosting!

The SBP 2011 Tutorial Presentations Are Now Online!

If you missed the three excellent tutorials that were delivered this past Monday, as part of the Social Computing, Behavioral-Cultural Modeling, and Prediction (SBP) 2011 Conference at the University of Maryland College Park, you may access the presentations online!

Tutorial 1: Toolkits for Computational Social Science: Using Honest Signals to Predict and Shape Human Responses

Alex `Sandy' Pentland, MIT

Tutorial Slides: funf Mobile Sensing System

Tutorial Slides: Influence Model

Tutorial 2: Social Network Analysis of Personal and Group Networks

Allen Tien, Medical Decision Logic

Chris McCarty, University of Florida Survey Research Center

Eric C. Jones, University of North Carolina-Greensboro

Tutorial Slides

Tutorial 3: Understanding Social Media: Tools, Applications, and Processes

Nitin Agarwal, University of Arkansas at Little Rock

Tutorial Slides

The above photos were taken at these tutorials.

Thanks again to the outstanding presenters and to the audiences for the great questions and discussions! Also, thanks to the tutorial presenters for making their lecture slides available!

Modern Economic Theory -- where it went wrong and how the crisis is changing it

When I travel there are two things that I never leave home without -- my BOSE headphones and the latest issue of The Economist. Many overseas flights have been more enjoyable and productive because of these two reliable "companions." My next trip, which will not require an airplane, however, will be to Cambridge (Massachusetts) to teach a course in Executive Education at Harvard University. The course is on portfolio optimization, clearly a very timely topic given the financial crisis. My lecture and handouts are prepared and I am ready for an exciting experience. I will be bringing, of course, my latest issue of The Economist (July 18-24, 2009), which is motivating, in part, this posting.

The cover of this issue of The Economist has a brown book with the title, "Modern Economic Theory," that is melting like chocolate in the summer heat. The caption reads: "Where it went wrong -- and how the crisis is changing it." The issue has several articles, which further reinforced the relevance of my my perspective on the material that I will be covering in my course at Harvard.

My view of portfolio optimization is that of a critical building block for general financial models that can capture complex interactions among agents in the economy. Classical portfolio optimization problems also have an elegant network structure that can be exploited both for conceptualization and visualization purposes and also for computation of solutions. Such building blocks can help in the construction of macroeconomic models in which one can then explore all sorts of policies such as taxes, tariffs, price supports, transaction costs, etc., and upon which one can also build international financial network models and even explore financial intermediation. I have been writing a lot on financial networks with intermediation and electronic transactions and have supervised doctoral dissertations on the subject.

Amazingly, in an article, in this issue of The Economist, on page 66, is written "In many macroeconomic models, therefore, insolvencies cannot occur. Financial intermediaries, like banks, often don't exist." Perhaps some of those macroeconomic modelers should read the literature! A great place to start would be the volume: "Innovations in Financial and Economic Networks" that I edited in 2003 and was published by Edward Elgar. Other relevant papers can be found on the Virtual Center for Supernetworks website.

Then, to add further injury, on page 67, and I quote a remark attributed to David Colander, who has been surveying economists: "Instead of solving models 'by hand', using economists' powers of deduction, he proposes simulating economies on the computer." The entire field of Computational Economics is over a decade old and those of use who work in the interfaces of operations research and economics and finance well know the power and the use of not only mathematical models to capture the intricacies of human and economic interactions but also the use of algorithms and computers to predict the results of such interactions, including product and financial flows and prices. We have a Society for Computational Economics, an annual conference (the most recent one just concluded in gorgeous Sydney, Australia), and even a journal!

There is a serious disconnect between the literature and the state-of-the-art and what some economists and policy makers are aware of. It is high time that areas of computational social sciences, including computational economics, get the recognition that they deserve! There is some hope, nevertheless -- on page 69 of the same issue of The Economist, Andrew Lo of MIT is quoted and his novel idea of creating a financial equivalent of the National Transport Safety Board to deal with future financial crises! Coincidentally, and you heard it here first, we will be hosting Professor Lo in our Fall 2009 INFORMS Speaker Series at UMass Amherst (co-sponsored with the Finance series). I can hardly wait!