Addressing adjacency constraints in rectangular floor plans using Monte-Carlo Tree Search

Abstract Manually laying out the floor plan for buildings with highly-dense adjacency constraints at the early design stage is a labour-intensive problem. In recent decades, computer-based conventional search algorithms and evolutionary methods have been successfully developed to automatically generate various types of floor plans. However, there is relatively limited work focusing on problems with highly-dense adjacency constraints common in large scale floor plans such as hospitals and schools. This paper proposes an algorithm to generate the early-stage design of floor plans with highly-dense adjacency and non-adjacency constraints using reinforcement learning based on off-policy Monte-Carlo Tree Search. The results show the advantages of the proposed algorithm for the targeted problem of highly-dense adjacency constrained floor plan generation, which is more time-efficient, more lightweight to implement, and having a larger capacity than other approaches such as Evolution strategy and traditional on-policy search.

[1]  Krishnendra Shekhawat,et al.  Introduction to generic rectangular floor plans , 2018, Artificial Intelligence for Engineering Design, Analysis and Manufacturing.

[2]  Eugénio Rodrigues,et al.  An evolutionary strategy enhanced with a local search technique for the space allocation problem in architecture, Part 1: Methodology , 2013, Comput. Aided Des..

[3]  Eugénio Rodrigues,et al.  An approach to the multi-level space allocation problem in architecture using a hybrid evolutionary technique , 2013 .

[4]  Fernando Marson,et al.  Procedural floor plan generation from building sketches , 2015, The Visual Computer.

[5]  Edwin Kinnen,et al.  Rectangular duals of planar graphs , 1985, Networks.

[6]  Sartaj Sahni,et al.  A linear time algorithm to check for the existence of a rectangular dual of a planar triangulated graph , 1987, Networks.

[7]  Demis Hassabis,et al.  Mastering the game of Go with deep neural networks and tree search , 2016, Nature.

[8]  George Stiny,et al.  Shape Grammars and the Generative Specification of Painting and Sculpture , 1971, IFIP Congress.

[9]  John S. Gero,et al.  Evolving design genes in space layout planning problems , 1998, Artif. Intell. Eng..

[10]  Sartaj Sahni,et al.  A linear algorithm to find a rectangular dual of a planar triangulated graph , 1986, 23rd ACM/IEEE Design Automation Conference.

[11]  Ipek Gursel Dino,et al.  An evolutionary approach for 3D architectural space layout design exploration , 2016 .

[12]  David Silver,et al.  Combining online and offline knowledge in UCT , 2007, ICML '07.

[13]  Abbas Afrazeh,et al.  A new model for dynamic multi floor facility layout problem , 2010 .

[14]  Ulrich Flemming,et al.  Representation and Generation of Rectangular Dissections , 1978, 15th Design Automation Conference.

[15]  Robert F. Woodbury,et al.  Whither design space? , 2006, Artificial Intelligence for Engineering Design, Analysis and Manufacturing.

[16]  William J. Mitchell,et al.  Optimal space planning in practice , 1981 .

[17]  Gabriela Celani,et al.  From the generation of layouts to the production of construction documents: An application in the customization of apartment plans , 2018, Automation in Construction.

[18]  Sergiu M. Dascalu,et al.  Towards creative design using collaborative interactive genetic algorithms , 2009, 2009 IEEE Congress on Evolutionary Computation.

[19]  Peter E.D. Love,et al.  Genetic search for solving construction site-level unequal-area facility layout problems , 2000 .

[20]  William J. Mitchell,et al.  Architecture's New Media: Principles, Theories, and Methods of Computer-Aided Design , 2004 .

[21]  Yavuz A. Bozer,et al.  Alternative approaches to solve the multi-floor facility layout problem , 1997 .

[22]  Keith C. C. Chan,et al.  EvoArch: An evolutionary algorithm for architectural layout design , 2009, Comput. Aided Des..

[23]  Krishnendra Shekhawat,et al.  Enumerating generic rectangular floor plans , 2018, Automation in Construction.

[24]  Kang Zhang,et al.  Customization and generation of floor plans based on graph transformations , 2018, Automation in Construction.

[25]  Nitant Upasani,et al.  Automated Generation of Dimensioned Rectangular Floorplans , 2019, Automation in Construction.

[26]  Demis Hassabis,et al.  Mastering the game of Go without human knowledge , 2017, Nature.

[27]  Robin S Liggett,et al.  Automated facilities layout: past, present and future , 2000 .

[28]  C. H. Antunes,et al.  Clustering of architectural floor plans: A comparison of shape representations , 2017 .

[29]  Marc Goetschalckx,et al.  Efficient Formulations for the Multi-Floor Facility Layout Problem with Elevators , 2007 .

[30]  Krishnendra Shekhawat,et al.  A Graph Theoretical Approach for Creating Building Floor Plans , 2019, CAAD futures 2019.

[31]  Richard S. Sutton,et al.  Reinforcement Learning: An Introduction , 1998, IEEE Trans. Neural Networks.

[32]  Eugénio Rodrigues,et al.  An evolutionary strategy enhanced with a local search technique for the space allocation problem in architecture, Part 2: Validation and performance tests , 2013, Comput. Aided Des..