Using Images to Detect, Plan, Analyze, and Coordinate a Smart Contract in Construction

Abstract

Reliable construction workflow relies on timely discovery, analysis, and checking of compliance with contract terms, which are time consuming and inefficient tasks. Smart contracts enabled by blockchain technology have demonstrated promise in addressing the inefficiencies of data communications due to their merits of traceability, immutability, transparency, and self-enforceability. However, a smart contract’s inability to interact with real-world data is the main issue that impedes further implementation. Today’s increasing availability of as-built data provides automatic condition assessments that have great potential to automate smart contract executions. This research area is uncharted territory for the industry. This research selects a case study to present an automatic decentralized management framework by exploring image-based deep learning solutions to automate and decentralize the conditioning of smart contract executions enabled by a web3.js-based decentralized blockchain application. It was found that the model can automate management intelligence with minimal workflow interruptions by timely identification of bottleneck activities and enforcement of mitigation strategies. Project managers can use the blockchain prototype to enhance information sharing, remove key risks, and enable a reliable workflow with minimal management efforts.