Article Preview
TopIntroduction
Managerial decision-making has always been a challenge. Finding the right course of action within a set of constraints constitutes one of the most important functions that a top executive must undertake. The level of complexity involved in decision-making increases substantially as globalization continues to develop worldwide. Corporate business schemes need to take into consideration regional conditions where manufacturing is performed; cross-country differences may influence overall decisions in terms of operations as well as financial results. Legal requirements and corporate policies play a key role when strategic planning points towards the economic well-being of the enterprise. The saying, “Think global, act local” (Geddes, 1915; Walter, 2004) is the rule to be followed today.
Data driven decision-making is being introduced in environments where intuition, common sense, and past experiences usually provide the basis for management alternatives to be chosen. Large amounts of data are stored in databases and data warehouses in order to be processed through the application of one or more scientific tools providing trends, confirming business hypotheses, and even presenting the most favorable scenarios of a company for top management from which to choose. The cycle that begins with Big Data (BD) then proceeding to Business Analytics (BA) and ending with a strategic decision as Business Intelligence (BI) is a common improvement procedure systematically applied by top business performers around the world and they make decisions based on rigorous analysis (MIT Sloan Management Review and IBM Institute of Business Value, 2010). Through such methodology, there is a continuous search for growth, efficiency, and competitive advantage. Business value is the reward obtained, but the primary focus is financial.
BA provides a rigorous, systematic, quantitative, and data driven approach to decision-making (Birnberg, 2009; Davenport & Kim, 2013). As a business enabler, BA has three types of analytics to be applied to the data collected: descriptive, predictive, and prescriptive (Bayrak, 2015). Prescriptive Analytics (PA) is the focus of interest with regards to planning. The question to be answered is: What is the best course of action to be taken, given a strategic economic objective to be achieved and subject to a set of operating constraints? Forecasting, cost estimation, and mathematical optimization are some of the tools that can help answer such question in a quantitative manner.
In fact, the application of powerful techniques of mathematical optimization from Operations Research (OR) is not new in business decision-making (Dechow & Mouritsen, 2005; Romanenko & Artamonov, 2014; Asllani, 2015). A comprehensive review is listed in Righetto et al. (2016). The need to incorporate financial aspects of a business enterprise into these tools has been indicated for more than half a century (Baumol, 1952; Vatter, 1967; Hartley, 1968); in particular, models that lead to optimization in policy and investment strategies have been recently proposed by Zhou et al. (2016) and Righetto et al. (2016).
Given a specific decision model, the likelihood of success of the solution suggested is always a function of the reliability of the data employed. The traditional algorithms that lead to a solution have been thoroughly tested and managers can depend on the use of OR tools constantly. However, frequent developments in the international business arena give rise to new problems with challenging objectives requiring extended models and methods (Merchant, 2012; Accenture, 2013; Kiron et al., 2014).