Decision models and artificial intelligence in supporting workforce forecasting and planning

For any organization, the effective workforce planning is essential to stay competitive and continue to subsist. Workforce planning is an organized process for identifying the number of employees, their mix and the types of skill sets required to accomplish an organization's strategic goals and objectives. This thesis focuses on demand analysis (i.e. forecasting the future workforce demand) in workforce planning. Workforce demand forecasting techniques can be classified into two broad categories viz. qualitative and quantitative. Generally, quantitative techniques are used to forecast workforce size and mix, whereas, qualitative techniques forecast competency requirements. This research explores demand analysis in many folds. First, state-of-the-art of workforce analysis techniques are presented and synthesized into a scenario specific forecasting technique(s) selection tree. Afterwards, the Clonal C-fuzzy Decision Tree (C2FDT), a decision support model, is proposed to forecast future workforce demand. C2FDT inherits its properties from fuzzy c-mean clustering and clonal algorithm. From the literature of workforce planning eight key parameters are selected as the major determinants of workforce analysis outcomes. In order to collect time-series and cross-sectional data corresponding to these parameters, set of questions are made. These questions are given to experts and according to their responses questions are integrated with the aid of Fuzzy Logic Controller. In this way large amount of data set is collected to train and test the C2FDT model.