The Internet of Responsibilities: tracking who is accountable with graphs and blockchain

Accountability is hard to manage in the workplace, especially for safety. A new paper proposes the "Internet of Responsibilities" (IoR), an accountability system built on big data and blockchain. The authors report that practical deployments showed an obvious increase in responsibility scores and safety awareness among personnel and organizations.

How the system works

The IoR model has four layers: perception, network, data and application. It lists the responsibilities of each position, with a focus on hazardous positions, and connects positions in directed acyclic graphs (DAGs) that show the hierarchy of responsibilities in the organization. The system collects responsibility data, marks risk areas by the positions of the responsibility nodes, and uses an automatic reminder and assignment system to enforce accountability.

Fig. 1: The Internet of Responsibility model, by the authors.

Fig. 2: Directed acyclic graph for responsibility lists, by the authors.

Scoring and sealing

Each person's responsibility score is built from a completion score and an execution score. The completion score totals 1000 points: by default 750 for daily work and 250 for additional work, a split that can be reset. It reflects whether staff submit the required materials on time to show a responsibility was completed. The execution score uses the submitted materials and co-workers' evaluations. Low scores trigger warnings to employees. Using blockchain, responsibility data is sealed and cannot be changed, so its authenticity and integrity are guaranteed, and accident accountability can be traced back through the weighted graph.

In practice

• Enterprises. The model was deployed at Hangzhou Quanshi Software Co., Ltd. A mobile app generates the tree of responsibility lists, staff submit data through it, and the system scores them automatically and ranks them within the department or organization.
• Government. For government accountability, the model shows safety evaluation scores from the sealed data, ranks safety level among units by responsibility score, and summarizes risk levels so units with falling scores are reminded at once. Risk shown by color, from red through yellow and green to blue, lets governments oversee their agencies with near real-time supervision.

Sources

• Xuejiao Tang, Jiong Qiu, Wenbin Zhang et al., The Internet of Responsibilities-Connecting Human Responsibilities using Big Data and Blockchain, arXiv:2312.04729 (2023), licensed CC0. Figures 1 and 2 are the authors' own. Changed here: summarized and reorganized, deployment screenshots and references left out; rewritten in hubnx's own words.