An integrated Cognitive Reliability and Error Analysis Method (CREAM) and optimization for enhancing human reliability in blockchain

Abstract

Minor errors in smart contract coding on the blockchain can lead to significant and irreversible economic losses for transaction parties. Therefore, mitigating the risk posed by coding errors is crucial, necessitating the development of approaches to enhance human reliability in coding. The Cognitive Reliability and Error Analysis Method (CREAM) is one such approach, examining how environmental conditions affect the human error probability (HEP). Within CREAM, Common Performance Conditions (CPCs) influence error probability. This study ranks CPCs in smart contract coding based on their importance in coding reliability using the Bayesian Best Worst Method (BWM). Two methods are developed based on basic CREAM. In the first method, experts specify the control mode based on their opinions, and the probability of experts’ coding errors is determined according to the control level. In the second method, an optimization problem is formulated to select the most suitable programs, enhancing experts’ coding reliability. The proposed model considers energy, cost, and organizational budget factors to identify the optimal smart contract while minimizing the risks and costs associated with human errors. A case study in the electronics supply chain validates the applicability and efficacy of the proposed methods. Results from the first method indicate an opportunistic control mode. In contrast, the proposed model shows that improving CPC levels has a more significant effect, shifting the control mode towards a tactical control and reducing HEP to 0.00249.