Client-centered detached modular housing: na tur al langua ge pr ocessing-enabled design recommender system

Abstract

Designing modular housing is a complex task that necessitates a thorough understanding of the di v erse needs of clients in terms of both design aesthetics and floor plan lay out. Furthermore , adhering to design for man ufactur e and assemb l y (DfMA) principles adds to the complexity, as these are essential modular home r equir ements. Traditional construction methods fr equentl y fail to meet the specific needs of both clients and DfMA, potentiall y r esulting in suboptimal design solutions. Incorporating client r equir ements during the design phase necessitates the use of an effecti v e system and fr amew ork to reduce changes in subsequent project stages. Existing liter ature lac ks a suita b le appr oach, particularl y in the context of modular housing. To address this gap, this paper introduces an artificial intelligence-building information modeling recommender system (RS) for detached modular housing design. The system pr ocesses client r equir ements enter ed as text utilizing the Word2v ec algorithm with the GloVe dataset, r efined thr ough transfer learning using surveyed client data of housing needs. The system r ecommends thr ee distinct modular building design alternati v es sourced from a building information modeling models database using cosine and Euclidean similarity functions. A sensitivity analysis ensures that client needs are considered fairly, increasing the robustness of the RS. By incorporating natural language processing, this system transforms the construction industry by making initial designs more client-centric compared with traditional methods. Furthermor e, it pr omotes impr ov ed colla boration among clients, design, and construction teams, reducing modifications to design in later stages of construction.