A Data-driven Human Responsibility Management System

Xuejiao Tang, Jiong Qiu, Ruijun Chen, Wenbin Zhang, Vasileios Iosifidis, Zhen Liu, Wei Meng, Mingli Zhang, Ji Zhang

Introduction

Workplace safety management often falters due to the inherent fallibility of human oversight. Traditional systems rely on manual monitoring, which struggles to scale alongside the complexity of modern corporate hierarchies. In large enterprises, the "supervision chain" frequently weakens as tasks are decentralized, making it difficult to pinpoint risk sources or verify adherence to safety protocols. This paper proposes a "smart" safety management system that transitions from subjective human engagement to an objective, data-driven approach capable of monitoring hazards in real-time and enforcing individual accountability.

System Architecture and Methodology

The proposed framework integrates IoT sensor data with organizational logic to automate safety supervision. The architecture is built upon three primary pillars: IoT integration, Responsibility Big Data, and Directed Acyclic Graphs (DAGs).

Figure 1: Conceptual framework mapping the transition from risk management to accident prevention through responsibility sets.

Responsibility Quantification and Modeling

The system defines "Boundaries of Responsibility" based on legal and professional standards, which are then translated into "Responsibility Sets"—specific tasks assigned to various departments. To manage complex organizational structures, the system utilizes DAGs to map relationships between responsibility nodes. This allows the system to programmatically define supervisory roles and task flows between superiors and subordinates.

Automation and Evaluation

A logic-based "Self-Reminder Module" monitors task attributes, such as periodicity. When high-risk deadlines approach or sensors detect hazards, the system automatically alerts relevant personnel. Performance is quantified through a "fulfillment score" derived from task completion data, IoT-verified evidence, and supervisory input.

Implementation and Key Results

The system was evaluated through a real-world implementation in a large shopping mall, utilizing IoT sensor streams from fire alarms and equipment monitors alongside manual fulfillment logs.

Figure 4: The mobile interface for the Daily Responsibility Reminder, showing task status and fulfillment scoring.

The practical application demonstrated significant improvements in management efficiency:

  • Emergency Automation: During fire alarm simulations, the system automatically notified patrols and logged response times without manual intervention.
  • Performance Visualization: Management could access "Personnel Responsibility Score Curves" to identify specific timeframes or departments where safety compliance fluctuated.
  • Accountability: By treating the hierarchy as a weighted DAG, the system ensures that if a subordinate fails a task, the supervisor is also alerted, maintaining the integrity of the safety chain.

Discussion and Future Work

While the system successfully automates many aspects of safety, it still faces a "human input" bottleneck where manual confirmation is required for certain tasks. Future work aims to integrate more diverse sensor data to reduce reliance on manual entries, particularly in high-stakes environments like the hazardous chemical industry. Additionally, the granular tracking of individual performance scores necessitates further exploration into workplace privacy and potential algorithmic bias.

Summary updated

Abstract

An ideal safe workplace is described as a place where staffs fulfill responsibilities in a well-organized order, potential hazardous events are being monitored in real-time, as well as the number of accidents and relevant damages are minimized. However, occupational-related death and injury are still increasing and have been highly attended in the last decades due to the lack of comprehensive safety management. A smart safety management system is therefore urgently needed, in which the staffs are instructed to fulfill responsibilities as well as automating risk evaluations and alerting staffs and departments when needed. In this paper, a smart system for safety management in the workplace based on responsibility big data analysis and the internet of things (IoT) are proposed. The real world implementation and assessment demonstrate that the proposed systems have superior accountability performance and improve the responsibility fulfillment through real-time supervision and self-reminder.
BibTeX
@inproceedings{tang2020data,
  author = {Xuejiao Tang and Jiong Qiu and Ruijun Chen and Wenbin Zhang and Vasileios Iosifidis and Zhen Liu and Wei Meng and Mingli Zhang and Ji Zhang},
  title = {{A Data-driven Human Responsibility Management System}},
  booktitle = {IEEE BigData},
  year = {2020},
  url = {https://openreview.net/forum?id=2RgWFb3NdT},
}
Figure