BEMPAS: A Decentralized Employee Performance Assessment System Based on Blockchain for Smart City Governance

Emmanuel Boateng Sifah, Hu Xia, Christian Nii Aflah Cobblah, Qi Xia, Jianbin Gao, Xiaojiang DuIEEE Access 2020 · DOI

Abstract

In the evolving landscape of Smart Cities, efficient governance relies heavily on the performance of government employees. However, traditional assessment systems often suffer from a "trust gap" due to human bias in manual reviews or the vulnerability of centralized digital systems to data tampering. This paper proposes BEMPAS, a decentralized framework leveraging a "3-chain" blockchain architecture and game theory to automate and secure employee performance evaluations. By utilizing Hyperledger Fabric, BEMPAS ensures transparency, immutability, and non-repudiation. Experimental evaluations show that the system's game-theoretic decision-making model outperforms traditional fuzzy logic in processing latency, making it a scalable solution for modern municipal administration.

Introduction

The governance of a Smart City requires a robust mechanism for monitoring and evaluating the performance of its workforce. Current evaluation processes typically fall into two categories: manual assessments, which are prone to favoritism and prejudice, or centralized digital systems, which present a single point of failure and are susceptible to administrative tampering. These issues lead to a significant trust gap, negatively impacting employee morale and organizational efficiency.

Smart City components highlighting the role of Smart Government.

There is a critical need for an automated, transparent, and immutable system that tracks behavior and awards credits without the risk of unauthorized data modification. BEMPAS addresses these challenges by moving the evaluation process to a decentralized ledger, ensuring that performance data is objective and tamper-proof.

Methods

The BEMPAS framework is built on Hyperledger Fabric and introduces a novel 3-chain model integrated into a single unified blockchain. This architecture separates distinct functional domains to enhance privacy and performance:

  1. ID-Chain: Manages digital identities by linking government IDs with biometric data (fingerprints, face, voice). This ensures non-repudiation, preventing employees from denying their actions.
  2. Behavior Chain: Acts as the data ingestion layer, collecting "on-the-job" data from Management Information Systems (MIS) and IoT sensors (GPS, RFID). It calculates a Participation Index (PI) based on observed behaviors.
  3. Credit Chain: Handles rewards and punishments using a game-theoretic approach to calculate payoffs, removing human discretion from the final evaluation.
The BEMPAS system architecture featuring the 3-chain blockchain model.

The system also incorporates a Key Management Center with Hardware Security Modules (HSM) for key protection and a Cloud Platform for storing large-scale heterogeneous data. To maintain blockchain performance, only data hashes and pointers are stored on-chain.

Experimental Evaluation

The authors conducted a Proof of Concept (PoC) using a desktop environment (Intel i-7, 16GB RAM) running Ubuntu 16.04 and Hyperledger Fabric v1.0.2. The simulation involved transactions for up to 100 employees and 100 blocks to assess processing overhead.

A key component of the evaluation was comparing the proposed game-theoretic decision model (seeking a Nash Equilibrium) against a traditional Fuzzy Logic approach. The results indicated that while Fuzzy Logic is common in HR, it creates complex structures that increase processing time as the system scales. In contrast, the game theory model proved computationally leaner and faster.

Key Results

  • Scalability: ID generation and verification times increased linearly with the number of employees. For 100 employees, ID generation was completed in approximately 190ms.
  • Processing Efficiency: Block processing time also scaled linearly, remaining manageable as the chain grew.
  • Algorithmic Superiority: During stress tests, the game-theory model consistently exhibited lower processing latency than fuzzy logic. As the employee count reached 100, the game theory model saved significant computation time per evaluation cycle.
  • Security: The system demonstrated robust resistance to data tampering through SHA-256 hashing and ensured that identity revocation required consensus among administrators.

Discussion and Conclusion

BEMPAS provides a decentralized solution to the trust gap in employee performance assessment. By automating evaluations through a 3-chain blockchain architecture, it ensures objectivity and security. However, the system faces challenges such as the high storage and power requirements of blockchain technology and the initial complexity of onboarding biometric and IoT infrastructure. Furthermore, the reliability of the system depends on the accuracy of the physical IoT layer, which remains susceptible to sensor spoofing. Future work will focus on integrating BEMPAS into broader Smart City domains and improving the resilience of the IoT data collection layer.

Summary updated

BibTeX
@article{sifah2020bempas,
  author = {Emmanuel Boateng Sifah and Hu Xia and Christian Nii Aflah Cobblah and Qi Xia and Jianbin Gao and Xiaojiang Du},
  title = {{BEMPAS: A Decentralized Employee Performance Assessment System Based on Blockchain for Smart City Governance}},
  journal = {IEEE Access},
  year = {2020},
  doi = {10.1109/access.2020.2997650},
}
Figure