An entropy based design evaluation model for architectural competitions through multiple factors

Architectural competitions offer architects the chance to present their design ideas, propose innovative solutions to design problems, and be rewarded for doing so. Architectural competitions try to set a competitive environment in which participants with different levels of experience can compete under equal conditions. This competitive environment encourages participating architects to improve the technical and aesthetic qualities of their projects. The design brief, while an important component of architectural competitions that outlines the general framework of the competition and guides the participants to some extent by laying down the rules, falls short of defining the complex relationships the buildings that are expected to be designed have. That the brief, being a written document, includes ambiguities and does not define well the design problems involved causes participating architects to have to interpret the brief and shape their designs based on their interpretation. Despite the brief being a document that outlines the general framework of the competition and is not amended in the course of the competition, it can be observed that neither the participants base their design, nor the jury members base their judgement solely on the brief. Considering the overall structure of architectural competitions, while the brief introduces objectivity to the evaluation process the evaluations of the members of the jury based on their personal experience and approaches add subjectivity. Generally, the evaluations in competitions are based on quality. In this context, there are significant debates on how quality is evaluated and the new meanings that quality might have in different contexts. Many criteria are involved in the evaluation of the project qualities in competitions. In addition to the criteria specified in the brief, many other inter-related criteria, identified by the members of the jury, emerge during jury evaluation. Hence, a great number of criteria play a role, with varying degrees of importance, in both the design and evaluation processes. For example, such criteria as the solid-void balance and functional distribution on the building plan are tackled by architects in the design process for the building and taken into consideration by the jury while evaluating the qualities of the design. The order of importance and weights of these two criteria in the design and evaluation phases are not fixed and differ according to the approaches of the designers and jury members. It is important to ensure fairness in the evaluation process so that the competitive environment, which is one of the fundamental characteristics of competitions, is not harmed and the confidence placed in architectural competitions is not eroded. Although the staging of competitions and the selection of jury members in a way that they will evaluate the projects in an unbiased manner are actions taken to ensure fairness in architectural competitions, problems might still occur in the evaluation processes in architectural competitions. Some of the various problems that might arise in the evaluation of projects in architectural competitions are as follows: not observing the competition procedures, administrators at different levels of hierarchy interfering with jury decisions, jury members carrying out their evaluations on anonymous projects by attempting to guess the identity of the designer, scoring being based on a limited number of criteria and excluding some criteria that could be of importance, and jury members trying to select the projects of the people who are alumni of the same university as they are. For the thesis, the results of jury evaluations, which can still suffer problems despite all the efforts to ensure fairness and which are ultimately subjective, have been analysed using an objective and calculated method. A model has been developed that questions the relationship between the results garnered through objective evaluation methods based on calculations of the measurable physical qualities of the projects (buildings) and the results (success ranking) of subjective jury evaluations of the same project (building). An entropy method based on the physical features of the participating projects has been used to reach objective and unambiguous results. Theoretical and practical studies have been carried out on the concept of entropy in physics, biology, ecology, economics, linguistics, art, and architecture. In architecture, there are many studies that have examined and found correlations between the levels of entropy values calculated based on various measurable physical features of buildings and subjective responses they make on people such as likability and preference. An in-depth analysis of the studies including entropy measurement in architecture has found several shortcomings. It has been found that the previous studies in the literature did not investigate architectural products with a high degree of complexity in terms of form and programme, conducted entropy measurements based on one physical feature of the building alone, and that those studies which did measure the entropy values of more than one physical feature simultaneously measured the entropy values in isolation, independent from one another. That is why in this thesis we have performed an entropy measurement of a customized architectural typology involving complex relationships and many physical features. Furthermore, we have developed a method that handles in an inter-related manner the entropy values calculated based on multiple factors. Municipal buildings composed of repetitive components, as in LEGO models, and involving functional variety was selected as the architectural typology on which to perform the entropy measurement. In addition, because of the existence of a large number of resulted architectural project competitions for municipal buildings, the project archive to be looked into was wide. 42 award-winning projects in 7 completed municipal building architectural competitions held from 2015 to 2017 in Turkey have been examined. Entropy values of 6 single factors (physical features) have been calculated by way of the entropy measurement methods developed as part of the thesis – namely, solid-void, outline, shape, functional distribution, spatial relation, and spatial flow. It has been estimated that factors should have weight coefficients on the assumption that all the factors affect the evaluation process at the same time but in different weights. The overall entropy, which can be obtained by the direct total sum of entropy values calculated for multiple factors should be valid when there is no relationship between factors. Considering that the factors are inter-related, weighted overall entropy values of the projects have been calculated by multiplying multiple factors with weight coefficients and summing the results. According to the results of the studies in the literature, there is a relationship between the entropy values of the buildings and the positive subjective impressions they make that can be explained by an inverted U graph. In the value range identified according to the relationship on the inverted U graph, the most average creates the most positive subjective effect. Like in the studies in the literature, it has been assumed that there would be an inverted U type relationship between the calculated Weighted Overall Entropy values of the award-winning projects and the success achieved by the projects in jury evaluations. Within the scope of the thesis, entropy values of the buildings have been measured based on single factors. It is necessary to ascertain the weight coefficients so as to calculate Weighted Overall Entropy values. At this point, weights have been calculated in such a way that the Weighted Overall Entropy values of the projects correspond to the positions they hold in the inverted U graph in line with their competition achievements. Weighted Overall Entropy values and weights have been calculated both on 6 projects by analysing the competitions one by one and all 42 projects by inserting them to optimization simultaneously. As a result of the thesis, Weighted Overall Entropy values based on the weights calculated for one competition and 7 competitions have been placed in an inverted U graph and the similarity and proximity of their positions to the results of the competitions have been examined. Furthermore, in the conclusion section, predictions have been provided about the areas and fields in which the developed model can be used and how it can be further improved.