An intelligent decision support system for workforce forecasting and planning




Tripathi, Mukul

Journal Title

Journal ISSN

Volume Title



This research examines, analyzes and proposes an effective integration of various artificial intelligence techniques as a forecasting tool in the workforce planning domain. In general, a specific workforce plan for an organization should include (a) a clear statement (quantitative or qualitative analysis) of what it is trying to achieve, (b) the level of detail it is focusing upon (how much workforce increase is needed or the mix of workforce needed), and (c) associated uncertainties with forecasting.

However, in practice, the implementation of workforce planning activities to achieve diverse objectives poses several interrelated and complex tasks. For example, these dissimilar tasks include collection of data from a diverse category through survey. Thereafter, the second challenging task is to filter the relevant information from the bulk data collected by categorization it into relevant factors (parameters) of interest.

On the other hand, there is general consensus for recognizing the workforce forecasting as a hybrid task of simultaneous time series and cross-sectional data analysis. Therefore, it is necessary to collect comprehensive information about a particular organization and conduct an extensive data analysis. The intelligent decision support system proposed in the research attempts to forecast future workforce for any organization. In this respect, a set of classification/forecasting techniques (e.g. neural networks, decision trees etc.) have been utilized for forecasting and their learning algorithm is evolved by utilizing the proposed Self-Guided-Ant based Genetically-Optimized-Neural-Network ( SGA GONN). This new algorithm is the core of the decision support system and is developed by utilizing several improved variants of neural network, genetic algorithm, and ant colony optimization. The performance of the proposed algorithm has been verified by comparing to conventional Neural Network, Genetic Algorithm Neural Network (GANN) and C2FDT (a previously developed algorithm) for prediction purposes.

This research also develops workforce forecasting software which is equipped with a highly customizable graphical user interface (GUI) developed in Visual C#. The GUI not only provides capabilities of customizing the algorithm and also helps in maintaining the database. Thus, users with minimum knowledge of GA and NN will still be able to operate the software, maintain the database and execute the algorithm by simply following the intuitive step by step GUI software.


This item is available only to currently enrolled UTSA students, faculty or staff. To download, navigate to Log In in the top right-hand corner of this screen, then select Log in with my UTSA ID.


Ant Colony Optimization, Artificial Intelligence, Data Mining, Neural Network, Software Development, Workforce Forecasting



Mechanical Engineering