Thursday, October 15, 2015

Data Science vs. Decision Science

Either of them is not the end, but the means to the end - for problem-solving, improvement, and innovation.

From a business perspective, "Data science is the study of where information comes from, what it represents and how it can be turned into a valuable resource in the creation of business and IT strategies"( Decision science or engineering is not "just a new buzzword." It is a knowledge revolution for proactive structural decision simulation analysis and strategic decision analysis for crisis early warning and proactive feedforward reactors control against process operation uncertainties. The typical challenge seen with the traditional analytics approach is to arrive at insights, but not necessarily affect actual business decisions, and not always in a timely manner. Decision Engineering approach embeds analytics in actual business decisions - rather than leaving it to the receiver of insights to use or dump. There is a further difference between Data Science and Decision Science. The latter is also called Business Data Science, combining the instrumental (data science tools & technique), social (business context) and functional (information process) axes to add value to the data and information within a company.

Data science has to be able to contribute towards the ability or capacity to enable change. Usually data-gathering is driven by the need to make decisions. Many are questioning the ROI on data science. But in the operational context of many organizations, data is collected as part of an important business function: human resources, logistics, quality control, and accounting. This means the data will be collected even if the ROI is not apparent. Strategic alignment, however, dictates constant change either in the nature of the data collected or its handling; This is not as simple as scanning all sorts of historical data that might now be outdated and inapplicable. While certain segments of the data science community might be focused on this historical learning, businesses generally seek out guidance to deal with future developments. So in certain respects, the only surviving lessons from the past are the most abstract; and the lessons also lead to the development of capacity to make use of intellectual capital; this remains with an organization in perpetuity, unless deliberately impaired. As such, a data scientist actually wields not just data but capital (capitalized intelligence) as evidenced by persistent artifacts associated with production.

There are three general pillars in Decision Science: Modeling/Simulation (stochastics and probability), Data Analysis (an application of statistics principles and data science), and Optimization (mathematical modeling with generally discrete results and an opportunity for sensitivity analysis for skeptics). These three pillars are different methods an analyst could apply to a particular problem based on the information at hand and the desired methodology. All three methods provide an opportunity to gain insight to make an informed quantitative decision and evaluate if a certain risk is acceptable. And there is an opportunity to explore a mixture of the "pillars," it doesn't matter if you are calling yourself a decision of data scientist, you are still applying scientific and mathematical principles to gain further insight or arrive at some results.

What is the goal of doing both data science and decision science? The pyramid of wisdom separate: data - information - knowledge - understanding - wisdom. This is in order of largest to smallest. These elements are not the same, but the one cannot be without the other. Nor is the one better nor equal they are all relevant and to be taken into account in decision making. Too often, businesses lean towards data science as the solution, building unbalanced teams which fall somewhat short of the fundamental skill of business problem definition and structuring. A solution is somewhere between data science and decision science. Businesses needs insights that drive real value, data science is only one of many enablers, and decision science could put more focus on decision analysis, but always keep in mind: either of them is not the end, but the means to the end - for problem-solving, improvement and innovation.


Post a Comment

Twitter Delicious Facebook Digg Stumbleupon Favorites More