Depicting Decision-Making: A Type-2 Fuzzy Logic Based Explainable Artificial Intelligence System for Goal-Driven Simulation in the Workforce Allocation Domain

• Published in: 2019 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE) pp. 1 - 6
• Main Authors: Ferreyra, Emmanuel, Hagras, Hani, Kern, Mathias, Owusu, Gilbert
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01-06-2019
• Subjects: Artificial intelligence; big bang-big crunch; Computational modeling; Explainable AI; Fuzzy logic; Fuzzy sets; goal-driven simulation; interval type-2 fuzzy logic systems; Resource management; Uncertainty
• ISSN: 1558-4739
• DOI: 10.1109/FUZZ-IEEE.2019.8858933

The recent years have witnessed a growing anticipation for the positive transformation of industries which adopt Artificial Intelligence (AI) for the core areas of their business activities. However, the effectiveness and reliability of such AI systems must comprise the ability to explain their data acquisition, the underlying algorithms operations and the final decisions to stakeholders, including regulators, risk managers, supervisors and end-users among others. There are plenty of areas where Explainable AI (XAI) holds the promise to be a major disruptor. Particularly, in Telecommunication Service Providers (TSPs) which is a core business activity relating to the workforce allocation domain, which, involves costly and time-consuming scheduling processes. This paper focuses on the construction of an XAI framework to assist workforce allocation based on a big bang- big crunch interval type-2 fuzzy logic system (BB-BC IT2FLS) for modelling and scaling goal-driven simulation (GDS) problems, specifically within the telecommunications industry. The obtained results reported the proposed XAI system produces similar results to opaque box models like Neural Networks (NNs) and LSTM Recurrent NNs while being able to explain the decision and operation of the employed system.