Dr. Amir MOSAVI, Ph.D. Data Scientist Technische Universität Darmstadt
Predictive Decision Making Amir Mosavi Abstract. Prediction brings insight into unknown. Accurate predictions can potentially transform businesses, industries, and almost any organization. Marketing, financial services, insurance, retail, and healthcare are just a few industries seeking for accurate predictions to enhance their decisions. Today, humanity more than ever seeks accurate predictions to better react to the climate changes. Due to uncertainties, complexity of the prediction functions and high computation costs, the conventional mathematical modeling approach cannot provide any reliable prediction model. Instead predictive analytics, emerged from data science, identifies patterns in big data to build predictive model for organizations. Although prediction modeling is meant to empower the decisionsupport systems, in reality it could not be beneficial without decision-analysis, as yet identifying the best possible response to the valid prediction is the actual problem that organizations facing now. In fact optimizing the optimal decisions and anticipation of every decision and its consequences must be also predicted and optimized. To scale to the complexity created Dr. Mosavi coins the term “predictive-decision model� a novel integration of prediction analytics with decision modeling, where predictions are optimized and an intelligent agent makes automated decisions relying on learning algorithms and decision preferences. This will revolutionize the way decision-support systems function today.
Dr. Amir MOSAVI, Ph.D. Data Scientist Technische Universität Darmstadt
Background The purpose of decision-support systems is to empower organizations facing complicated and large-scale problems. Decision-support systems function based on decision analysis which means evaluating the alternatives' characteristics on a number of criteria or attributes in order to rank, sort or choose among the alternatives. The complexity of real-life decision making problems originates from the presence of multiple decision criteria. Typically, there does not exist a unique optimal solution for the real-life problems. Thus it is necessary to use preferences to differentiate between solutions. Structuring complex problems in the context of multiple decision criteria explicitly leads to more informed and better decisions. However there is not a unique optimal solution to a multiple criteria decision problem that can be obtained without incorporating preference information. The solution typically requires a series of mathematical programming models in order to reveal implicitly defined solutions. And for the mathematical programming models containing integer variables with substantial computational difficulty multiple objective optimization algorithms are used. Since the start of the modern multiple-criteria decision-making discipline in the early 1960s numerous approaches and methods introduced. However among the vast number of methods introduced none can actually anticipate the consequences of the decision made. In the other words decision support systems are not equipped with the tools to predict the consequences associate with a solution. In fact prediction research and decision-making research communities have been often apart. Integration of decision-analysis tools with predictive analytics; although it seams to be a powerful concept for creating the next generation of decision-support systems, it has never been the case. Nevertheless decision-support systems can highly benefit from the latest technological advancements of predictive analytics. However it is clear that predictive
Dr. Amir MOSAVI, Ph.D. Data Scientist Technische Universität Darmstadt
analytics has not contributed to the decision-analysis systems as it probably should have. To integrate decision-making and prediction, Dr. Mosavi coins the term “predictive-decision model� what he believes in its extensive potential for transforming latest discoveries into a major innovation in decision science and prediction.
Details of Research Prediction brings insight into unknown. However the true purpose of prediction is about taking an action with more knowledge about the consequences of it. Accurate predictions can potentially transform businesses, industries, and almost any organization. Marketing, financial services, insurance, retail, election forecasting and healthcare are just a few industries seeking for accurate predictions to enhance their decisions. Today, humanity more than ever seeks accurate predictions to better react to the climate changes. Due to uncertainties, complexity of the prediction functions and high computation costs, the conventional mathematical modeling approach cannot provide any reliable prediction model. Instead predictive analytics, emerged from data science, identifies patterns in big data to build predictive model for organizations. Although prediction modeling is meant to empower the decisionsupport systems, in reality it could not be beneficial without decision-analysis, as yet identifying the best possible response to the valid prediction is the actual problem that organizations facing now. Predictive analytics cannot be beneficial without existence of decision analysis. In fact a prediction is successful when it leads to the optimal decision and consequently right action. Taking the right action is as important as making the accurate prediction. Although predictive analytics
Dr. Amir MOSAVI, Ph.D. Data Scientist Technische Universität Darmstadt
provide powerful tools for prediction, the organizations are left with making decision without an insight into possible choices. Not knowing that the consequences of every decision that may be made is not clear. The reality is more complicated than that. To make an informed decision the consequences of possible decisions also must be predicted. Organization in addition to the prediction models, are to be provided with the consequences of every single decision. In fact optimizing the optimal decisions and anticipation of every decision and its consequences must be also predicted and optimized. To scale to the complexity created Dr. Mosavi coins the term “predictive-decision model� a novel integration of prediction analytics with decision modeling, where predictions are optimized and an intelligent agent makes automated decisions relying on learning algorithms and decision preferences. This will revolutionize the way decision-support systems function today. Dr. Mosavi will conduct research on the integration of two research fields of prediction and decision-making applying this concept in a variety of reallife problems. His vision is to create complete models of predictive-decision to be able to make informed and automated decisions. He proposes an integrated system of predictive analyics and decision-analysis where numerous predictions are analyzed for identifying the optimum decision leading to the right action.
A. MOSAVI Dr. Amir Mosavi, Ph.D. Darmstadt 15 December 2015
Dr. Amir MOSAVI, Ph.D. Data Scientist Technische Universität Darmstadt
“Dr.Mosavi received numerous research awards, four postdoc fellowships, and managed more than 15 international research stays at the top research institutes. He is the Green-talent awardee of 2015 praised by German research minister for his promising research on predictive models for climate change risk reduction. In 2013 he was among the top five scientists in the world to win the UNESCO-TWAS for his work of accurate prediction-decision tools for reducing the climate change impacts. As an expert in both data science and decision science, he consulted world's largest companies with advancement of their prediction and decision models to anticipate the consequences of the potential actions in order to make better decision in an automated manner. Dr.Mosavi coins the term “predictivedecision model” what he believes in its extensive potential for transforming latest discoveries into a major innovation in decision science and prediction.” ~ A. MOSAVI. 2015.