Kanalalterungsmodelle, mit denen sich der Zustand von Abwasserkanälen simulieren lässt, können wertvolle Werkzeuge für die Sanierungsplanung sein. Dennoch werden sie in Deutschland bisher nur von wenigen Kanalnetzbetreibern eingesetzt. Im Rahmen des Forschungsvorhabens SEMA-Berlin wurden verschiedene Modellansätze getestet und hinsichtlich ihrer Prognosequalität bewertet. Für den Modellaufbau wurden die Ergebnisse von mehr als 100 000 TV-Inspektionen sowie Daten zu den individuellen Kanaleigenschaften und Umgebungsfaktoren der Stadt Berlin verwendet. Die Untersuchungen zeigen, dass das statistische Modell GompitZ die Zustandsverteilung des Kanalnetzes mit einer Genauigkeit von 99 % wiedergeben kann. Mit Random Forest, einem Modell des maschinellen Lernens, kann mit einer Trefferquote von 67 % vorhergesagt werden, welcher Kanal sich im schlechten Zustand befindet. Die Ergebnisse können dafür genutzt werden, prioritäre Haltungen für Kanalinspektionen zu identifizieren und Investitionen so zu steuern, dass der Zustand der Kanalisation langfristig erhalten oder sogar verbessert wird.

Deterioration models can be successfully deployed only if decision-makers trust the modelling outcomes and are aware of model uncertainties. Our study aims to address this issue by developing a set of clearly understandable metrics to assess the performance of sewer deterioration models from an end-user perspective. The developed metrics are used to benchmark the performance of a statistical model, namely, GompitZ based on survival analysis and Markov-chains, and a machine learning model, namely, Random Forest, an ensemble learning method based on decision trees. The models have been trained with the extensive CCTV dataset of the sewer network of Berlin, Germany (115,258 inspections). At network level, both models give satisfactory outcomes with deviations between predicted and inspected condition distributions below 5%. At pipe level, the statistical model does not perform better than a simple random model, which attributes randomly a condition class to each inspected pipe, whereas the machine learning model provides satisfying performance. 66.7% of the pipes inspected in bad condition have been predicted correctly. The machine learning approach shows a strong potential for supporting operators in the identification of pipes in critical condition for inspection programs whereas the statistical approach is more adapted to support strategic rehabilitation planning.

UV/Vis spectrophotometers have been used for one decade to monitor water quality in various locations: sewers, rivers, wastewater treatment plants (WWTPs), tap water networks, etc. Resulting equivalent concentrations of interest can be estimated by three ways: i) by manufacturer global calibration; ii) by local calibration based on the provided global calibration and grab sampling; iii) by advanced calibration looking for relations between UV/Vis spectra and corresponding concentrations from grab sampling. However, no study has compared the applied methods so far. This collaborative work presents a comparison between five different methods. A Linear Regression (LR), Support Vector Machine (SVM), EVOlutionary algorithm method (EVO) and Partial Least Squares (PLS) have been applied on various data sets (sewers, rivers, WWTPs under dry, wet and all weather conditions) and for three water quality parameters: TSS, COD total and dissolved. Two criteria (r2 and Root Mean Square Error RMSE) have been calculated - on calibration and verification data subsets - to evaluate accuracy and robustness of the applied methods. Values of criteria have then been statistically analysed for all and separated data sets. Non-consistent outcomes come through this study. According to the Kruskal-Wallis test and RMSEs, PLS and SVM seem to be the best methods. According to uncertainties in laboratory analysis and ranking of methods, LR and EVO appear more robust and sustainable for concentration estimations. Conclusions are mostly independent of water matrices, weather conditions or concentrations investigated.

Deterioration modelling has been developed in the last decades to support operators and municipalities in defining mid-long term asset management strategies with limited availability of sewer condition data (CCTV). Modelling can help validating and showing the viability of current strategies or provide information to justify the relevance of additional investments and expenditures. Several modelling approaches are now available but not commonly used by sewer operators and municipalities to support strategies mainly because of the lack of real scale demonstration of the tangible benefits provided. Indeed, most of these models fail to show that they can adequately forecast future conditions (Ana and Bauwens, 2010; Scheidegger et al., 2011; WERF, 2012).

Deterioration modelling can be a powerful tool to support utilities in planning efficient sewer rehabilitation strategies. However, the benefits of using deterioration models are still to be demonstrated to increase the confidence of utilities toward simulation results. This study aims at assessing the performance of a statistical deterioration model to estimate the current condition and predict the future deterioration of the network. The quality of prediction of the deterioration model GompitZ has been assessed using the extensive dataset of 35,826 inspections of the city of Braunschweig in Germany. The performance of the statistical model has been compared with the performance of a simple model based only on the condition of observed sewers. Results show that the statistical model performs much better than the simple model for simulating the deterioration of the network. The findings highlight the relevance of using modelling tools to simulate sewer deterioration and support strategic asset management.

Im Rahmen des Forschungsprojekts SEMA ist die Prognosequalität eines Alterungsmodells anhand der TV-Inspektionsdaten der Stadt Braunschweig geprüft worden. Die Qualität der Prognose wurde auf der Grundlage einer Probe von 35.826 Inspektionen bewertet. Die Inspektionen wurden mittels eines substanzbasierten Modells klassifiziert. In einem zweiten Schritt wurde das statistische Modell KANEW-Z angewandt, um die Kanalalterung zu simulieren. Der Vergleich der Inspektions- mit den Simulationsergebnissen zeigt, dass das Modell in der Lage ist, die Zustandsverteilung des Systems ziemlich genau wiederzugeben. Die Ergebnisse sind auch ermutigend auf individueller Haltungsebene. Im Allgemeinen zeigt das Alterungsmodell viel bessere Ergebnisse als ein einfaches lineares Alterungsmodell. Schlussfolgernd unterstreichen die Ergebnisse das Interesse und den potentiellen Nutzen der Anwendung von Alterungsmodellen zur Unterstützung von Asset-Management-Strategien.

This paper reports about experiences gathered from five on-line monitoring campaigns in the sewer systems of Berlin (Germany), Graz (Austria), Lyon (France) and Bogota (Colombia) using UV-VIS spectrometers and turbidimeters. The influence of local calibration on the quality of on-line COD measurements of wet weather discharges has been assessed. Results underline the need to establish local calibration functions for both UV-VIS spectrometers and turbidimeters. It is suggested to practitioners to calibrate locally their probes using at least 15-20 samples. However, these samples should be collected over several events and cover most of the natural variability of the measured concentration. For this reason, the use of automatic peristaltic samplers in parallel to on-line monitoring is recommended with short representative sampling campaigns during wet weather discharges. Using reliable calibration functions, COD loads of CSO and storm events can be estimated with a relative uncertainty of approximately 20 %. If no local calibration is established, concentrations and loads are estimated with strong errors questioning the reliability and meaning of the on-line measurement. Similar results have been obtained for TSS measurements.

Several infrastructure studies highlight the ongoing deterioration of critical assets in water and wastewater systems (WERF, 2007). A recent survey among 397 water and wastewater industry participants in the U.S.A. and Canada highlights that aging infrastructure and the management of capital and operational costs are the two main industry issues (Black and Veatch, 2013). From the participants, more than 70% of municipalities and utilities have already implemented condition assessment and inspection programs to assess the condition state of their systems. However, less than 10% are currently using simulation tools to support their asset management strategies. These results underline the strong opportunity for municipalities and utilities to increase the efficiency of their asset management programs by extracting the value of their (already) available data. Several modeling approaches are now available but not commonly used by sewer operators to support strategies (Caradot et al., 2013). Indeed, most of these models still fail to show that they can adequately forecast future conditions (Ana and Bauwens, 2010; Scheidegger et al., 2011). This article presents an assessment of the ability of sewer deterioration models to simulate the condition distribution of sewer networks. The analysis has been done using the extensive CCTV dataset of a German city, Braunschweig.

This article presents an assessment of the quality of prediction of a Markov-based statistical sewer deterioration model using the extensive CCTV dataset of a German city, Braunschweig. Additionally, a sensitivity analysis has been performed in order to assess the influence of input data availability on model performance. Results indicate that models are able to simulate quite accurately the condition distribution of the network with deviations smaller than 1%. Results also indicate that the performance of deterioration models is quite independent of the amount of CCTV data available to calibrate the model. Even when using very few data (˜3%, i.e. 1000 inspections) to calibrate the model, very good model performance can be obtained.This article presents an assessment of the quality of prediction of a Markov-based statistical sewer deterioration model using the extensive CCTV dataset of a German city, Braunschweig. Additionally, a sensitivity analysis has been performed in order to assess the influence of input data availability on model performance. Results indicate that models are able to simulate quite accurately the condition distribution of the network with deviations smaller than 1%. Results also indicate that the performance of deterioration models is quite independent of the amount of CCTV data available to calibrate the model. Even when using very few data (˜3%, i.e. 1000 inspections) to calibrate the model, very good model performance can be obtained.

Asset management is an increasing concern for wastewater utilities and municipalities. Sewer deterioration models have been developed by research and municipalities to support the definition of cost-effective inspection and rehabilitation strategies. However, the acceptance of deterioration models among sewer operators and decision makers still raise considerable challenges. This article presents the state of the art of condition classification and sewer deterioration modeling and discusses key issues for the future development of deterioration models. Research is needed (i) to identify the most appropriate approaches for condition classification and deterioration modeling and (ii) to conclude clearly about their quality of prediction. Due to the high costs associated with CCTV inspection and data collection, the influence of input data on modeling quality and the optimal input data requirement are still to be evaluated. The ongoing project SEMA aims precisely to assess the suitability of models to simulate sewer deterioration. Objectives and strategy are shortly presented at the end of the article.