The extraction of regulatory information is a prerequisite for automated code compliance checking. Although a number of machine learning models have been explored for extracting computer-understandable engineering constraints from code clauses written in natural language, most are inadequate to address the complexity of the semantic relations between named entities. In particular, the existence of two or more overlapping relations involving the same entity greatly exacerbates the difficulty of information extraction. In this paper, a joint extraction model is proposed to extract the relations among entities in the form of triplets. In the proposed model, a hybrid deep learning algorithm combined with a decomposition strategy is applied. First, all candidate subject entities are identified, and then, the associated object entities and predicate relations are simultaneously detected. In this way, multiple relations, especially overlapping relations, can be extracted. Furthermore, nonrelated pairs are excluded through the judicious recognition of subject entities. Moreover, a collection of domain-specific entity and relation types is investigated for model implementation. The experimental results indicate that the proposed model is promising for extracting multiple relations and entities from building codes.

An Ant Colony-based Heuristic Algorithm for Joint Scheduling of Post-earthquake Road Repair and Relief Distribution

Many construction conflicts can arise from the separation of design and construction in the current AEC industry. Constructability analysis has been proposed as an effective method to fill the gap between the AEC project design and construction. However, the project information model can be improved to integrate product information and process information and to provide a framework for incorporating construction experience and knowledge. The product oriented scheduling technique (POST) is presented to provide an alternative scheduling technique with the aim to integrate the static product system and the dynamic construction progress configuration in a uniform description context. Another aim of POST is to provide a knowledge management tool for construction sequence knowledge. This paper introduces the concepts and the network structure of the POST model. It also explores the incorporation of construction sequence knowledge into the component state network. It is suggested that the POST model can reduce errors arising from mapping product models and process models, and improve constructability analysis.

Scheduling play an important role in the approach from hazards transiting to accidents. This study explores the evaluation of safety constraints in terms of hazards, especially their temporal and spatial attributes, and further associates these two attributes with duration and workspace of construction activities, respectively. Moreover, a spatio-temporal analysis framework has been developed for semantically representing the necessary conditions for accident occurrence in terms of temporal and spatial relationships between hazard objects and accident victims. Based on these conditions, guidelines for rescheduling construction have been proposed in order to achieve safer work environment. This analysis framework is also validated against a case study of constructing the structure of a resident building.