Removing unexplained variability RCI
The project improves the measurement of road pavement conditions in Switzerland with enhanced accuracy, repeatability and reproducibility. We solve challenges such as data variability and incomplete data to enable better forecasting.
Factsheet
- Schools involved School of Architecture, Wood and Civil Engineering
- Institute(s) Institute for Building Materials and Biobased Products IBBM
- Research unit(s) Road Construction Materials Group FGS
- Funding organisation Schweizerische Eidgenossenschaft (Bundesverwaltung)
- Duration (planned) 01.08.2022 - 31.03.2025
- Head of project Prof. Aybike Öngel
-
Project staff
Marcel Abegglen
Joël Lenk - Keywords Road condition measurement, standards, GIS, road construction, Big Data
Situation
The monitoring of the condition of Swiss roads is based on a system of detailed standards and guidelines, including SN 640 510 Properties of road surfaces, VSS-40511 Texture, VSS-40512 Skid resistance measurements, VSS-40517 Longitudinal evenness, VSS-40518 Transverse evenness and VSS-40525B Requirements. The data collected is important for predicting the condition, determining action strategies and defining specific measures. It is crucial for this data to be accurate, comprehensible, repeatable and reproducible. In recent years, the use of automated measuring equipment has improved the speed and precision of data collection. However, the data collected shows a high degree of unexplained variability. Variations can be caused by repeatability, reproducibility and other general uncertainty factors; these uncertainties can be mitigated by improving the physical measurement processes and the measurement methods. Possible causes of unexplained variability in the measurement system of Swiss road operators include: insufficient precision of detailed standards, inherent variability of measurement equipment/procedure (e.g. material, equipment, operators), external measurement conditions, different methods for the same measurement, underlying assumptions in measurement principles and procedural differences in data evaluation.
Course of action
The COEUS project focuses on improving the accuracy, comprehensibility, repeatability and reproducibility of road condition measurements in Switzerland. The aim of the project is to obtain more accurate estimates with respect to road deterioration, optimum intervention strategies and the right timing for specific measures. To achieve this goal, the project has several sub-goals. The first is to illustrate the extent of the unexplained variations in the existing data by analysing the historical data in accordance with existing practice. The second sub-goal is to identify the sources of variability and the impact of each, by generating and testing different hypotheses. The third sub-goal is to propose ways to filter out variability in the existing data; and the fourth is to propose ways to reduce variability going forward. The COEUS project has the potential to significantly improve the accuracy and reliability of road condition measurements in Switzerland. By eliminating the causes of variability and reducing their impact, the project will result in more accurate and usable data for decision-makers responsible for maintaining and improving the country’s road infrastructure.
