Search

PurposeContractors must check the provisions that may cause disputes in the specifications to manage project risks when bidding for a construction project. However, since the specification is mainly written regarding many national standards, determining which standard each section of the specification is derived from and whether the content is appropriate for the local site is a labor-intensive task. To develop an automatic reference section identification model that helps complete the specification review process in short bidding steps, the authors proposed a framework that integrates rules and machine learning algorithms.Design/methodology/approachThe study begins by collecting 7,795 sections from construction specifications and the national standards from different countries. Then, the collected sections were retrieved for similar section pairs with syntactic rules generated by the construction domain knowledge. Finally, to improve the reliability and expandability of the section paring, the authors built a deep structured semantic model that increases the cosine similarity between documents dealing with the same topic by learning human-labeled similarity information.FindingsThe integrated model developed in this study showed 0.812, 0.898, and 0.923 levels of performance in NDCG@1, NDCG@5, and NDCG@10, respectively, confirming that the model can adequately select document candidates that require comparative analysis of clauses for practitioners.Originality/valueThe results contribute to more efficient and objective identification of potential disputes within the specifications by automatically providing practitioners with the reference section most relevant to the analysis target section.

When bidding on construction projects, contractors need to understand the specifications properly to manage project risks. However, specifications are mainly analyzed based on human cogniti...

Field engineers should manually analyze the site appropriateness of every provision in a construction specification by comparing the requirements against the national standards. To support the manual review involving multiple documents, the authors proposed a semantic text-pairing method that identified relevant provisions from different specifications considering the textual properties. First, 2527 provisions were prepared from two construction specifications of highway projects undertaken in Qatar and five national standards from Australia, the United Kingdom, and the United States. Second, the Doc2Vec model trained the provisions and learned the textual features based on Paragraph Vector with Distributed Memory. Third, the provision relevance was estimated by normalizing cosine similarities between provision vectors generated by the Doc2Vec model. The experiments returned promising results, with an average matching accuracy of 84.40%. The results contribute to the specification review by automatically identifying the most relevant provisions and making the process objective and robust to human errors.