Skip to main content



Designing with Graph Theory

During our discussions in lecture regarding networks, we used something called graph theory to help understand these networks properly. We started by talking about the network of computer systems that essentially made up the Internet in the early 1970’s with graphs showing how each was connected. Other examples included social networks, credit card transactions and medical/blood donation. Overtime, we began to learn how graph theory is not just applicable in technological settings but in many aspects of our everyday lives as well, serving as a tool to process information and connect it appropriately. We used graph theory to help view networks and understand relationships between objects and people. But how else can we use graph theory to help improve our lives?

Dan Shipper is an Internet Entrepreneur who lives in Manhattan, NY. While on a subway ride to Brooklyn, Dan had a simple question: How could you build a better train? Initially, Shipper starts out with different designs he would use in designing the cars on the new train and than began to think about infrastructure development issues (ex: How many stations to create, where to put them etc., I don’t want to spend more than few dollars on a trip, I don’t want to carry around cash- the possibilities are endless.) Subsequently, Shipper begins to realize that in designing such a complicated system, the requirements needed to create to create an ideal train system are enormous. So many design problems could develop, how could you solve them all?

Shipper eventually came up with a list of 5,000 potential problems in designing his new train network. He called each individual problem a “misfit.” In order to resolve each of these individual problematic design pieces, using graph theory, Shipper took each of these individual misfits, decided which ones he felt were highly interdependent on each other and grouped and connected them together. After analyzing the resulting graph, Shipper decided which groups were critically important began to build features that would alleviate his concerns. He found that if you corrected misfits that maintained a high amount of connections (read highly related) to other misfits, solving the issues causing the highly connected misfits would also help solve the neighboring/connected misfits.  For example, misfits like “I don’t want to be hit by a train” or “I don’t want to worry about falling into the gap” and similar misfits would be connected to misfit “I want to feel safe.” If you work on a design problem of making everyone feel safer, by extension, you help solves other problems/misfits as well.

Granted, Shipper is not planning on building a transportation network anytime soon (he was just demonstrating a proof of concept). However, it is fascinating to see how graph theory can be used in practical applications outside the classroom and how it can be used to help solve read world issues like product and network design. Using graph theory, Shipper took what would be a drawn out and time consuming process of research and development and was able to create an effective method to streamline and improve design of large-scale system networks. If we can use graph theory to do something like build a next generation transportation network, imagine what else we can do.

Article: http://danshipper.com/building-a-better-train-to-brooklyn

Comments

Leave a Reply

Blogging Calendar

September 2015
M T W T F S S
 123456
78910111213
14151617181920
21222324252627
282930  

Archives