A novel framework for TER allocation using multilayer perceptron and intuitionistic fuzzy Z numbers for talent management

The effective allocation of Training and Education Resources (TER) is one of many organizational pathways to maximizing workforce capability and employee development. Conventional means of assessing employees and allocating employee resources are inadequate in managing uncertainty, imprecision, or performance data in complex forms and paradigms. In this paper, a new model is proposed that implements an integrated application of Intuitionistic Fuzzy Z-Numbers and multi-layer perceptron networks for a more realistic and accurate employee performance evaluation and resource allocation. The proposed model employs fuzzy logic to handle uncertainty in performance evaluation, such as degrees of membership, non-membership, and hesitancy. The multi-layer perceptron network predicts employee performance trends to help allocate resources, if required, while performance is progressing. The model was analyzed through experimental analysis, with a significant R² factor value (0.9967). The R² proves that the model predicts performance and improves employee resource allocation distribution. The proposed model is a demonstrative improvement in employee performance evaluation tools, compared to traditional frameworks of evaluation and allocation. The model is flexible enough to help organizations conduct effective talent management and allocate resources, with a handling degree of uncertainty, when their available employee performance data is incomplete. However, this framework should be explored further in terms of effective models that reduce data sparsity as well as real-time integrations and adjustments. Ultimately, this research presents an adapted and scalable model of organizational talent management and organizational performance.