04-06-2017 04:22 PM
As part of my MS in Analytics course curriculum, recently I had an opportunity to discuss about effective visualization of numeric data. I wanted to share it with sas community.
When we are planning to visualize static or past data, histograms, pie charts, scatter plots, bubble plots, correlation matrix etc. works for conveying numeric data. However, if we are working on hour-by-hour data and if we need to make contingency plans, in scheduling equipment, buses, flights etc. these charts fall short.
In such cases, if multiple variables need to be represented and there needs to be quick management action, animation that represents equipment, people, and effect of delay of the whole situation works best. The following link provides the animation mentioned here.
In the example provided, a bus is delayed and people wait at bus stop(s) and as more people wait, the delay gets further. If there are only a few buses, the delay would be longer. If we need to communicate this situation to management, animation with people, buses, and bus route would work best to get their buy-in for additional resources such as buses, employees (budget) and so forth.
Recommendations for improvement: When the dollars and cents are depicted of losses or the impact to the community due to bus delays, it would be much more effective to influence city / town leadership to enlist their sponsorship for additional budget. Further investigation could be done on this chart for use in factories, airlines, or wherever there is equipment, resource, and people involved.
Yau, N. (2013). Data points: Visualization that means something. 38-41, Indianapolis, IN: Wiley. ISBN: 9781118462195.
Kirk, A. (2012). Data Visualization: A successful design process. Olton, England: Packt.
Anonymous (n.d.) Why do buses bunch?
04-06-2017 05:51 PM
@Murali11: I totally agree! Just the other day someone reminded me of a presentation I participated in at a MWSUG meeting a couple of years ago. It was about a data mining problem that I was able to solve by using a visualization of the data technique, namely watching a map of the US midwest population over time as incidents of fraud were being detected. see: http://www.sascommunity.org/wiki/Expert_Panel_Solution_MWSUG_2013-Tabachneck
04-06-2017 08:40 PM
Mr. Art, I like the powerpoint presentation and the illustration with Art's Kansas Fried Chicken. The problem that you have mentioned is an interesting one in the sense that it is a true "needles in haystack(s) situation".
Thank you for sharing it.