Application of BIM Tools in Parametric and Generative Design for the Conception of Complex Structures

The present research applies the Parametric Design (PD) and Generative Design (GD) for the generation of complex structures, through the BIM methodology, being implemented in design phase of a new modern proposal for Pavilion J1 of the National University of Engineering from Perú. The research aims to: Study the PD and GD considering the interoperability provided by BIM tools, propose procedures that help solve PD and GD problems, understand the benefits of process automation through generative and parametric algorithms. The conception and design phase of projects are developed in a traditional way using CAD Softwares for drawing plans or BIM Softwares for the design and/or modeling of structures, carrying out manual tasks either for the extraction of measurements, exchange of information or modeling, this implies a lack of efficiency in many processes because despite having modern computational tools, the full potential they offer is not used. This is reflected in the productivity of the construction sector as it is one of the lowest compared to other sectors such as manufacturing, commerce, agriculture. Due to this problem, new technologies were studied, such as evolutionary algorithms supported by parametric design for the conception and design of structures. Subsequently, as a test, this new methodology was applied to various types of structures, testing the parametric behavior and understanding the operation of these new methodologies. As a result of the previous tests, key procedures were defined to cover parametric and generative problems, developing algorithms in textual code (Python), visual algorithms and applying generative algorithms (NSGA-II); capable of creating structures automatically adapting to the designer's criteria. Based on the last stage of the PD and GD procedures, the algorithms for the formulation of the structure were implemented in Block J1, demonstrating the applications and benefits in various tasks such as modeling, loads generation, structural design and software interoperability.

GenUI: interactive and extensible open source software platform for de novo molecular generation and cheminformatics

Many contemporary cheminformatics methods, including computer-aided de novo drug design, hold promise to significantly accelerate and reduce the cost of drug discovery. Thanks to this attractive outlook, the field has thrived and in the past few years has seen an especially significant growth, mainly due to the emergence of novel methods based on deep neural networks. This growth is also apparent in the development of novel de novo drug design methods with many new generative algorithms now available. However, widespread adoption of new generative techniques in the fields like medicinal chemistry or chemical biology is still lagging behind the most recent developments. Upon taking a closer look, this fact is not surprising since in order to successfully integrate the most recent de novo drug design methods in existing processes and pipelines, a close collaboration between diverse groups of experimental and theoretical scientists needs to be established. Therefore, to accelerate the adoption of both modern and traditional de novo molecular generators, we developed Generator User Interface (GenUI), a software platform that makes it possible to integrate molecular generators within a feature-rich graphical user interface that is easy to use by experts of diverse backgrounds. GenUI is implemented as a web service and its interfaces offer access to cheminformatics tools for data preprocessing, model building, molecule generation, and interactive chemical space visualization. Moreover, the platform is easy to extend with customizable frontend React.js components and backend Python extensions. GenUI is open source and a recently developed de novo molecular generator, DrugEx, was integrated as a proof of principle. In this work, we present the architecture and implementation details of GenUI and discuss how it can facilitate collaboration in the disparate communities interested in de novo molecular generation and computer-aided drug discovery.

Model Operations: Morphogenesis and the Design Process

Since the emergence of digital design techniques in combination with so-called responsive materials, the concept of organic forms in architecture seems to be gaining a new quality. The resemblance to an organism should no longer apply only superficially but be inscribed in the materiality as well as in the history of origin and functioning. This article addresses these new transformative effects between architecture and biology. They are presented primarily in relation to the structural architecture of the 1960s and the computational architectural systems since the 1990s. One focus of architecture is on dynamic forms that adapt themselves to their environment by means of flexible materials and generative algorithms. Here, architecture as technically animated matter no longer involuntarily competes with creative nature but is seen as part of a reciprocal relationship. This reciprocal relationship is specified by recourse to various architectural models. The models’ approaches suggest that organic-looking forms are generated by simulated biological processes. The article examines this claim of the models from the perspective of the history of architecture and design. It shows how, since the mid-twentieth century, a renewal of architectural design practice has been sought by reformulating morphological questions at the intersection of biological and cybernetic discourses.

WOOD AND GENERATIVE ALGORITHMS FOR THE COMPARISON BETWEEN MODELS AND REALITY

This study examines the emblematic case of a test room and its relation to digital modelling. This space is the result of a multi-optimization process that has been physically built for the verification of the initial hypotheses. As a result, it is actually a Physical Twin, designed to be transformable by removing a wall. The same space, on the other hand, has become useful for testing the Digital Twin logic by associating a BIM model with a dynamic representation of the data captured by the sensors. The representation is thus placed at the core of this cyclic phase between reality and representation, with the goal of validating the proposed theories through empirical practice, improving digital computational ability, and identifying pathways for monitoring space's interactions with the environment and those who live in it.

Invariant Representation of Physical Stability in the Human Brain

Successful engagement with the world requires the ability to predict what will happen next. Here we investigate how the brain makes the most basic prediction about the physical world: whether the situation in front of us is stable, and hence likely to stay the same, or unstable, and hence likely to change in the immediate future. Specifically, we ask if judgements of stability can be supported by the kinds of representations that have proven to be highly effective at visual object recognition in both machines and brains, or instead if the ability to determine the physical stability of natural scenes may require generative algorithms that simulate the physics of the world. To find out, we measured responses in both convolutional neural networks (CNNs) and the brain (using fMRI) to natural images of physically stable versus unstable scenarios. We find no evidence for generalizable representations of physical stability in either standard CNNs trained on visual object and scene classification (ImageNet), or in the human ventral visual pathway, which has long been implicated in the same process. However, in fronto-parietal regions previously implicated in intuitive physical reasoning we find both scenario-invariant representations of physical stability, and higher univariate responses to unstable than stable scenes. These results demonstrate abstract representations of physical stability in the dorsal but not ventral pathway, consistent with the hypothesis that the computations underlying stability entail not just pattern classification but forward physical simulation.

Attention-based generative models for de novo molecular design

Attention mechanisms have led to many breakthroughs in sequential data modeling but have yet to be incorporated into any generative algorithms for molecular design. Here we explore the impact of...