[Back]

A. Mahdavi, L. Berberidou-Kallivoka:
"Recent Advances in Open Daylighting Simulation Environments";
Talk: The 1995 IESNA Annual Conference, New York; 1995; in: "Proceedings of the 1995 IESNA Annual Conference", (1995), 408 - 418.

In earlier studies, the "two-way" inference approach and the notion of an "open" simulation environment were introduced as new paradigms for computer-aided daylighting modeling (Mahdavi 1993, Mahdavi and Berberidou-Kallivoka 1993, 1994, 1995 and Berberidou-Kallivoka 1995). These approaches involve not only the evaluation of a given design solution in terms of its performance, but also aim at deriving formal implications of performance criteria. Since the performance-to-design mapping is not a "unique" process (for example, there may be different room and window designs that would provide the same average illuminance), a preference-based formalization of evaluation criteria for design elements and performance indicators was introduced to facilitate the dynamic control of the different degrees of freedom of relevant design variables during the interactive design process.

The objective of this paper is to summarize recent advances toward the comprehensive implementation of these concepts into a new version of the program prototype GESTALT (Generative Simulation Tool for Archictectural Lighing). This new GESTALT version can operate both in an "explicit" mode using a fast-response computational module and in an "implicit" mode using a comprehensive simulator and investigate projection techniques. The applicability of the integration of an advanced daylighting simulation tool in the GESTALT environment is demonstrated. This enhanced daylighting simulation module allows for the consideration of sky models of arbitrary luminance distribution and a more resolved, radiosity-based, treatment of light inter-reflections. The functionality of the system is demonstrated through examples of computationally generated design variations based on daylight-related performance criteria.