OPTIWAND - Inverse Optimierung von Lärmschutzwanddimensionen

This project develops a method to find the optimal solution for a noise barrier plan regarding financial costs and effect. Inverse Optimization is minimizing an individual weighted costfunction consisting of barrier costs, violation of noise exposure limits, and additional constraints by modifying the barrier geometry. The mathematical algorithm takes into account the Austrian guidelines, standards, and legal regulations for street- and railway noise. The functionality is demonstrated with specific practical examples.

Planning of noise barriers for existing and new ASFINAG and ÖBB projects, as well as for street administration in Austrian provinces is based on defined guidelines (including recommendations, legal regulations and standards). The aim of these regulations is to fulfil noise exposure limits taking into account technical and economical constraints. The current procedure is an iterative forward planning method, where height and length of noise barrier segments are stepwise varied until the user detects an acceptable solution based on the given constraints. The finally reached result is very depending on the experience and
creativity of the individual planner. In general this iterative process is linked to an
overestimation on the needed barrier area and/or is not using the full potential of the invested financial resources to protect the neighbours in the most optimal way.

In this research project a method for inverse planning to be used in software solutions is developed, which is adapted to the special Austrian regulations to be used for ASFINAG and ÖBB projects. This method searches inversely the optimal noise barrier geometry based on a set of objectives. Such algorithms are currently used in different technologies and science (e.g. optimizing weights of components in aircraft design, radiotherapy, process optimization). Within a FFG funded research project the basic requirements to use inverse optimization in noise control have been analysed. In addition to the noise exposure limits also the costs for barriers, special protective windows, homogeneity of barrier heights and
special designs are taken into account. Each constraint can be weighted by individual factors, which can be set to achieve reproducible results. Especially individual buildings or parts of buildings (e.g. upper floors) can be extracted, where the noise exposure could also be limited by sound insulating windows. The economic efficiency can be increased substantially.

The results of this research project is the development of the mathematical basis for optimization based on Austrian regulations, the combination with existing experience in inverse planning for noise barriers and the development of a prototype to demonstrate the potential to plan the economical best, but also on the same side for the neighbours most optimal noise barrier solutions.


Application screenshot(s):