Sewer blockages can cause overflows and flooding, with consequences such as damage to property and environmental pollution, risks to public health and economic loss. Despite the causes being understood, blockages in sewer networks may occur unpredictably. The responsible staff for sewer networks at water utilities need to efficiently determine the most effective action (what), the specific network location needing attention (where), the optimal timing for intervention (when), and the appropriate remedial task (how), especially given the unpredictability of blockages. Today a reactive approach to asset management and maintenance is often adopted. Additionally, data availability, quality and interoperability between systems are not always at levels that can support decided objectives, proactive maintenance planning and asset management of pipe networks. Thus, the aim of this thesis is to propose and evaluate approaches that can support analytics-driven maintenance planning and asset management for sewer networks. These approaches aim to contribute to mitigating the impact of siloed data structures and enhance the understanding of blockage root causes from a spatial perspective.

In this thesis, the challenges of data management in the asset management of pipe networks were investigated through focus group workshops and questionnaire surveys. A conceptual framework was developed based on findings from focus group workshops and surveys. The framework combines data quality assessments, interoperability evaluations between asset management tools, data collection, and informational benefits analysis. This framework aimed to identify the presence of data silos and plausible pathways towards more data-driven data management strategies. A performance assessment combining performance indicators associated with blockages and partial least squares regression (PLS) was conducted to draw inferences that could be useful at a strategic level. Furthermore, a spatial heterogeneity assessment of blockages and factors affecting blockages was carried out. This approach combined network kernel density estimation (NKDE), network k-function, and geographically weighted Poisson regression (GWPR). Lastly, a vulnerability assessment was carried out that combined topological analysis using edge-based centrality measures and network cross-k-function. These approaches were applied to three sewer networks.

The focus group workshops and questionnaire surveys identified several challenges affecting data management in the context of pipe network asset management. Many of the challenges could be ascribed to issues related to data quality and interoperability. Results from the preliminary application of the conceptual framework showed how it could be applied for identifying data silos and pathways to data-driven decision-making towards proactive management blockages in sewers. The observed spatial trends and patterns from network k-function analysis and network kernel density estimation showed spatial variability in the occurrence of blockages (single occurring and recurring). Geographically-weighted Poisson regression analysis showed spatial heterogeneity in factors influencing blockage propensity. The network cross-k-function analysis demonstrated that pipes with historical blockage incidents tend to be clustered around critical pipes with higher centrality values. These results could support vulnerability assessments in sewer networks and the development of targeted maintenance strategies. These approaches together could aid data-informed maintenance planning and asset management at the strategic, tactical and operational levels.

Avloppsstopp kan orsaka översvämningar vilket kan medföra konsekvenser som skador på egendom och spridning av föroreningar och leda till ekonomiska förluster samt risker för folkhälsan. Trots att orsakerna till avloppsstopp är kända, inträffar ibland oförutsedda avloppsstopp i ledningsnäten. De som ansvarar för avloppsledningarna behöver kunna fastställa effektiva åtgärder (hur), den specifika punkt i nätverket som behöver åtgärdas (var) och den optimala tidpunkten för åtgärd (när). Idag sker ofta åtgärder på ledningsnät reaktivt. Vidare är inte alltid data för analys tillgänglig eller av god kvalitet. Det finns även problem med interoperabilitet mellan system som kan stödja uppsatta mål för proaktiv underhållsplanering och tillgångsförvalting av ledningsnät. Därför är syftet med denna avhandling att föreslå och utvärdera metoder som kan stödja analytiskt driven underhållsplanering och tillgångsförvalting av avloppsledningsnät. Dessa metoder syftar till att minska negativa effekter av så kallade datasilos och förbättra förståelsen av grundorsaker till avloppsstopp utifrån ett spatialt perspektiv.

I denna avhandling undersöktes utmaningar med datahantering inom tillgångsförvalting i ledningsnät genom fokusgruppsworkshopar och enkätundersökningar. Ett konceptuellt ramverk utvecklades, baserad på resultat från workshoparna och en av enkäterna. Ramverket innefattade användning av datakvalitetsbedömningar och utvärdering av interoperabilitet mellan verktyg för tillgångsförvalting och datainsamling. Vidare analyserades fördelar med att identifiera förekomsten av datasilor och tänkbara tillvägagångssätt för att nå mer datadrivna strategier för datahantering. En prestandautvärdering som kombinerade prestandaindikatorer relaterade till avloppsstopp som kombinerades med regressionsanalys med minsta-kvadrat-metoden genomfördes för att dra slutsatser som kan vara av nytta på strategisk nivå. Vidare gjordes en spatial bedömning av avloppsstoppens förekomst och deras variationer över ledningsnätet samt faktorer som påverkar avloppsstopp. Denna analys kombinerade en täthetsanalys (kernel density estimation) med en network k-function samt en geografiskt viktad Poisson-regression analys. Slutligen genomfördes en sårbarhetsbedömning som kombinerade en topologisk analys baserad på grafteori och network cross-k-function. Metoderna tillämpades på tre avloppsnät.

Fokusgruppsworkshoparna och enkätundersökningar identifierade en mängd utmaningar som påverkar datahantering i tillgångsförvalting av ledningsnät. Flera av utmaningarna kunde tillskrivas problem relaterade till datakvalitet och interoperabilitet. Resultaten från den preliminära tillämpningen av det konceptuella ramverket visade hur det kunde användas för att identifiera datasilor. De observerade spatiala trenderna genom analysen med network k-function och täthetsanalysen visade spatial variabilitet i förekomsten av avloppsstopp. Den geografiskt viktade Poisson-regressionsanalysen visade spatial heterogenitet i faktorer som påverkar förekomsten av avloppsstopp. Analysen med Network cross k-function visade att ledningar med historiska incidenter med avloppsstopp tenderade att grupperas kring rör med högre centralitetsvärden som med det bedömdes som mer kritiska för ledningsnätens funktion. Dessa metoder skulle tillsammans kunna bidra till en mer datainformerad underhållsplanering och tillgångsförvaltning på strategisk, taktisk och operativ nivå.

This study involved a survey of Swedish water utilities to evaluate their pipe-network data-collection objectives, usage, storage, and exchange routines. Factors impacting data integration (and the associated benefits) were also identified. Results showed that current data storage and exchange routines can be augmented to support commonly identified objectives and data utilisation needs, especially in larger water utilities. Levels of awareness of the opportunities for and benefits gained through asset management processes and data integration varied between utilities. Further research on the benefits of data integration in pipe network asset management is required to develop an evidence base on benefits accrued in practice, especially considering metadata, the diversity of legacy systems still in operation, costs and policy use.

In this study, the spatial relationship between critical pipes identified using edge-based centrality measures and pipes with higher failure probability-based vulnerability indicators were analysed in sanitary sewer networks. By analysing two sub-networks, one residential and the other a central network, significant spatial associations between pipes with high centrality values and those exhibiting adverse conditions (poor CCTV grades, previous blockages, and low self-cleaning capabilities) were identified. Path-based centrality measures, particularly edge betweenness and K-path edge centrality were less influenced by weights when identifying critical pipes. In contrast, non-path-based measures like nearest neighbour edge centrality could identify localised spatial patterns between critical pipes and pipes in adverse conditions within the sewer networks investigated. The results showed that the spatial patterns between critical pipes and pipes in adverse conditions were not random and could support proactive maintenance planning and the development of more resilient networks. Additionally, the impact of network structure, connectivity, and differences in the composition of pipe attributes could contribute to variations in the strength of observable spatial associations.

The topology of Sanitary Sewer Networks (SSNs) can play an influential role in the occurrence and magnitude of operational failures such as blockages and basements flooding (Reyes-Silva et al., 2020). For example, meshed (grid-like) topologies are reported to be less vulnerable to failures compared to branched (tree-like) topologies (Zhang et al., 2017). However, in reality, most SSNs are reported to have a predominantly branched topology (Reyes-Silva et al., 2020). Therefore, it could be argued that the spatial behaviour of operational failures may be related to the topological properties of SSNs. This study explored this argument by investigating the spatial association between the location of recurrent blockages and the location of influential nodes within the network. Graph theory–centrality measures (Ganesan et al., 2020) and the network cross-K-function (Okabe and Sugihara, 2012) were the methods used. Secondly, the question of which structural, hydraulic or environmental factors may explain the identified spatial associations was also explored using geographically weighted regression (Fotheringham and Charlton, 2009). In lieu of robust properly calibrated hydraulic models, results from centrality measures and network cross k-function can support the discovery of influential locations within the topology of SSNs that may propagate recurrent blockages. Such influential locations may also be inception points for remedial actions such as redesigning, which may be more cost-effective in the long term compared to conventional approaches like flushing. Results from a preliminary application of centrality measures and network cross-function to the SSN of one municipality (total network length 500 km, ≈40 people/km) using its historical blockage data are presented.

Blockages have been reported to account for a significant proportion of reported failures in sewer networks. The malfunctioning of the sewer network from blockages and the subsequent disruption to other public services and flooding may constitute a risk to the environment and human health. Due to the complex nature of underground sewer networks, a reactive approach to blockage maintenance is typically employed. However, although proactive maintenance strategies have been developed, both approaches could be expensive and highlight the need to address the problem with analytics-based methods. Although blockage triggering mechanisms may be known, sewer blockages often appear at random. Thus, it is necessary to improve the understanding of the influential mechanisms involved in forming blockages in sewer networks to support its maintenance and guarantee adequate performance levels. The overall aim of this thesis was to contribute with new knowledge, approaches and methods that can support improved proactive maintenance planning of blockages in sewer networks.

Various methods to achieve the aim have been investigated in relation to asset management planning levels. At the strategic level, blockages and associated performance indicators were employed in conjunction with Poisson and partial least squares regression to assess the performance of sewer networks, including gaining additional insights. At the tactical and operational levels, a procedure was developed. The procedure combines network k-function, geographically weighted regression and random forest ensembles. The network k-function analysis explains the significance of the spatial variation of blockages. The Geographically weighted Poisson regression (GWPR) investigates the degree of influence of explanatory factors on increased blockage propensity in differentiated segments of the sewer networks. Thirdly, the random forest ensembles was used to predict pipes with blockage recurrence likelihood. A proposed conceptual framework was applied at all asset management levels to assess the state of data-driven integrated asset management (IAM), based on data quality assessments, interoperability evaluations between IAM tools, and data collection and informational benefits analysis. 

Results from demonstrating the methods with data from the Swedish waters statistical database and three Swedish municipal sewer networks, namely A, B and C, are presented. Blockage related performance indicators showed that the average blockage rate in medium sized networks was 2-3 times the rate in other sewer networks in Sweden. Furthermore, sewer maintenance strategies were suspected to be ineffective, and increased proactive strategies may improve maintenance efficiency. The procedure in networks A, B and C indicated that the clustering of recurrent blockages maybe linked to an increased need for flushing-related maintenance in sewer pipe networks. The degree of influence between investigated factors and increased blockage propensity indicated that these relationships were not global (not the same in all locations) within and between the sewer networks for networks A, B and C. These non-stationary relationships were observed to occur in various forms, i.e. adequate self-cleaning velocity showed positive and negative correlations in different locations. The networks with relatively more substantial spatial clusters of blockages, higher data quality and availability were observed to have a higher mean prediction accuracy. The applied conceptual framework showed that intuitive asset management characterised the current state of blockage management in the municipal sewer network C with medium to good data quality and low interoperability.

Analytical tools used in infrastructure asset management of urban water pipe networks are reliant on asset data. Traditionally, data required by analytical tools has not been collected by most water utilities because it has not been needed. The data that is collected might be characterised by low availability, integrity and consistency. A process is required to support water utilities in assessing the accuracy and completeness of their current data management approach and defining improvement pathways in relation to their objectives. This study proposes a framework to enable increased data-driven asset management in pipe networks. The theoretical basis of the framework was a literature review of data management for pipe network asset management and its link to the coherence of set objectives. A survey to identify the current state of data management practice and challenges of asset management implementation in five Swedish water utilities and three focus group workshops with the same utilities was carried out. The main findings of this research were that the quality of pipe network datasets and lack of interoperability between asset management tools was a driver for creating data silos between asset management levels, which may hinder the implementation of data-driven asset management. Furthermore, these findings formed the basis for the proposed conceptual framework. The suggested framework aims to support the selection, development and adoption of improvement pathways to enable increased data-driven asset management in municipal pipe networks. Results from a preliminary application of the proposed framework are also presented.

Efficient management of sewer blockages requires increased preventive maintenance planning. Conventional approaches to the management of blockages in sewer pipe networks constitute largely unplanned maintenance stemming from a lack of adequate information and diagnosis of blockage causative mechanisms. This study mainly investigated a spatial statistical approach to determine the influence of explanatory factors on increased blockage propensity in sewers based on spatial heterogeneity. The approach consisted of the network K-function analysis, which provided an understanding of the significance of the spatial variation of blockages. A geographically-weighted Poisson regression then showed the degree of influence that explanatory factors had on increased blockage propensity in differentiated segments of the sewer pipe network. Lastly, blockage recurrence predictions were carried out with Random Forest ensembles. This approach was applied to three municipalities. Explanatory factors such as material type, number of service connections, self-cleaning velocity, sagging pipes, root intrusion risk, closed-circuit television inspection grade and distance to restaurants showed significant spatial heterogeneity and varying impacts on blockage propensity. The Random Forest ensemble predicted blockage recurrence with 60–80% accuracy for data from two municipalities and below 50% for the last. This approach provides knowledge that supports proactive maintenance planning in the management of blockages in sewer pipe networks.

Sewer pipe networks are expected to operate with minimal or no interruptions. The complex nature of randomlyoccurring failures in sewer networks arising from blockages significantly adds to the cost of operation and maintenance.Blockages are significant due to sewage backup or basements flooding, resulting from theiroccurrence. Therefore, continuous performance assessment of sewer pipe networks is necessary to ensurerequired levels of service at an acceptable cost. This study provides insight into the performance of the sewerpipe networks by assessing the proneness of the network to blockages. Furthermore it draws inferences at a holisticstrategic level of influential explanatory factors of blockage proneness, using data available in the SwedishWater and Wastewater Association’s benchmarking system. Results indicate that medium sized municipalitiesare prone to at least 30% more blockages per km per year compared to other municipalities. A hypothesis ofexplanatory factors includes reduced flow volumes and flow depth. Flow velocities below self-cleaning velocityin sewer pipe networks, encouraged by sluggishness of flow are responsible for increased possibility for sedimentdeposition and accumulation in sewers leading to blockages. This is also exacerbated by the deposition of nondisposables(wet wipes, baby diapers, hard paper, etc.), accumulation of fats, oils and grease in sewers andincreased water conservation measures.

Sewer networks are expected to operate with minimal or no interruptions. The complex nature of randomly occurring failures in sewer networks arising from blockages significantly adds to the cost of operation and maintenance. Blockages are significant due to sewage backup or basements flooding, resulting from their occurrence. Continuous performance assessment of sewer pipe networks is necessary to ensure a required level of service at an acceptable cost. This study provides insight into the performance of the sewer pipe networks by assessing the proneness of the network to blockages and drawing inferences at a holistic strategic level of influential explanatory factors of blockage proneness, using data available in the VASS statistics database managed by Swedish Water and Wastewater Association. Results indicate that medium sized municipalities are prone to at least 30% more blockages per km per year compared to other municipalities. A hypothesis of explanatory factors includes, reduced flow volumes and design flow depth resulting in flow velocities below self-cleaning velocity in sewer pipe networks encouraging the sluggishness of flow and increased possibility for sediment deposition and accumulation in sewer leading to blockages. This is also exacerbated by the deposition of non-disposables (wet wipes, baby diapers, hard paper etc.), accumulation of FOGs in sewers and increased water conservation measures. Blockages also show a trend indicating occurrence may have a constant return frequency in medium – less than small sized municipalities.