One of the key changes associated with the adoption of digital technology is its capacity to measure employees' workloads in an agile, data-driven manner and to optimise workforce planning. Telecommunication companies in Indonesia face increasingly complex challenges in managing workforce agility amid evolving roles, skill requirements, and business portfolios. However, existing workload analysis (WLA) methods remain largely manual, limiting their scalability, responsiveness, and accuracy. This study aimed to develop a machine learning (ML)–based WLA model using the full-time equivalent (FTE) approach, with a particular focus on its application in the Indonesian telecommunications sector. Structured data relevant to WLA calculations were extracted from the company's D-mobile application system, a platform used for daily operational reporting and employee activity tracking. The findings identify five dominant factors influencing WLA scores within the telecommunications sector: cultural assessment, job position, take-home pay, most recent promotion, and functional assessment results. The ‘stretch’ category exhibited the highest proportion of individual FTE scores. This study contributes to the development of ML-based WLA models, particularly in telecommunications contexts, and offers practical implications for enhancing agile workforce management in the digital era.
Wahyuningtyas et al. (Fri,) studied this question.