Advances in artificial intelligence (AI) and machine learning (ML) present opportunities to advance architectural spatial layout planning (SLP) and address limitations of previous research. Architects practicing SLP have long prided themselves on a deep understanding of topological spatial relationships, spatial geometry, and aesthetics. SLP remains a fundamental measure of design quality, which typically refers to more subjective attributes such as aesthetics, emotional impact on the environment and people interacting with it, and well-functioning layout. [1]. Traditionally, architects and designers relied on sketches on tracing paper to create his various layout options. This allows you to deal with programmatic relationships in 2D and 3D. [2]. However, this process is iterative and labor-intensive, and conflicting constraints and time pressures often require problem-solving skills rather than creative thinking. [3]. Often the best way to resolve an SLP is to generate a variety of alternatives until most team members agree on the chosen option in a reasonable manner. For example, abductive reasoning or logical reasoning that architects used to do on a daily basis based on the most input. Probably an explanation. As a result, computational assistance has become invaluable to architects and designers, especially when tackling repetitive and complex tasks that are unsuited to creative professionals. Automatic spatial layout planning (ASLP) has emerged as a solution to these nonlinear design problems. [4] Since the mid-20th century, it has evolved through automation, composition, planning, and allocation. [5].
ASLP is developed using various methodologies such as shape grammars, genetic algorithms, evolutionary algorithms, and case-based design. [6]. These generative design methods have been particularly successful in engineering fields where the metrics are relatively simple in terms of weight, strength, and intensity. [7]. However, traditional ASLP has complex relationships and trade-offs between parameters, which requires excessive time and effort to find the optimal design solution. [8]. Moreover, resilience and reusability remain important challenges when applying generative design methods to the ASLP domain. [9]. Furthermore, traditional methods may not fully consider all relevant factors or adapt to changing requirements such as building codes and site conditions. These limitations can result in a suboptimal final design that does not fully meet your design requirements. [10].
Recently, the new computational revolution, AI, has come under intense scrutiny from all angles and is already impacting many different industries at an unprecedented pace, including engineering, physics, finance, marketing, and real estate. Masu. [11]. Similarly, AI can significantly contribute to ASLP by removing responsibility, automating mundane tasks, and supporting decision-making processes. [12]. Additionally, generative ML creates unexpected designs through accumulated or newly generated data. All of this technological advancement and potential is a promising research avenue in the rapidly expanding field of AI. [13]And it encourages designers to rethink the means of acting as creative writers in the midst of upheaval. Therefore, research on her ASLP using AI is needed to clarify the extent to which this implementation impacts architectural practice. Designers need to leverage AI approaches to find the semantics of nondeterministic polynomial-time complete (NP-C) problems. [4] Evaluate approaches while understanding explicit and implicit properties [14].
This paper focuses on the use of AI techniques in ASLP, the most popular among ML and its subsets Deep Learning (DL) and Reinforcement Learning (RL), and discusses the challenges, tasks, and future in various contexts. Consider it comprehensively and holistically. A systematic view. It should be noted that inherent human constraints limit bibliographic analysis. This leads to the temptation to replace quantitative estimates as unbiased scientific properties with definitive descriptions. [15]. For example, a relatively small number of published journal articles can have a large impact on a research field. However, due to the limited number of available papers, quantitative analysis may not be able to accurately identify specific research topics. Therefore, a qualitative review was comprehensively performed to enhance the quantitative analysis and enlighten specific topics.
Thus, this study uses a mixed review approach quantitatively and qualitatively to examine previous studies and discuss research themes and scope through a systematic perspective of ASLP+AI. This method combines both qualitative and quantitative methods to thoroughly explore and group related topics based on data analysis. Therefore, quantitative analysis (i.e., bibliographic analysis) aims to identify research goals in the ASLP+AI field by analyzing existing research using relevant keywords and discussing knowledge gaps and challenges. The purpose is Additionally, qualitative reviews (i.e., systematic reviews) attempt to establish connections with the architectural design field by performing in-depth analyzes and identifying possible avenues for future research. This is accomplished by identifying research gaps and providing recommendations in the future directions section.
The structure of this document is as follows. This document begins by reviewing the terminology used in the field of ASLP+AI. The methodology section then provides an overview of the methodology, including bibliographic analysis using text mining and the criteria used to collect data for the study. Finally, a systematic review in the detailed discussion section delves into the fundamental components of AI applications in the areas of image-based, graph-based, performance-based, and agent-based approaches. In this paper, we use AI as a broad term that includes traditional ML, DL, and RL statistical models. This paper aims to provide designers with a comprehensive resource on the field of ASLP+AI by identifying challenges, gaps, and future research areas based on ASLP and its AI methodologies. The scope of this study focuses on examining peer-reviewed journal and conference papers from the past decade that address ASLP using AI techniques.
