Peter Chalk Centre

University of Exeter

Stocker Road

Exeter

EX4 4QD

Tel: +44 (0)1392 263637

E-mail: CCWI2019@exeter.ac.uk 

17th International Computing & Control for the Water Industry Conference

1st - 4th September 2019
University of Exeter, UK
  • Facebook Social Icon
  • Twitter Social Icon
  • Instagram

8C Asset management

Please note the time shown on this page is automatically adjusted by the server according to the time zone set in your computer.

Collaborative

Zheng Wu

Chair:

to

Optimal Water Pipe Replacement Scheduling Based on Life Cycle Cost Evaluation and Mathematical Optimization

Shinsuke Takahashi

Presenter:

Authors:

Shinsuke Takahashi and Hiroyuki Ohta

Efficient management of water infrastructure assets is one of the most significant challenges in the water industry. Efficient replacement scheduling of pipelines, which account for a high proportion of water assets, is required to reduce future investment. We propose optimal water pipe replacement scheduling based on a life cycle cost(LCC) evaluation and mathematical optimization. The LCC for each pipe attribute is calculated as a sum of the costs for pipe replacement, repair, leakage loss, and so on. The base year for replacement (the optimal replacement timing) is determined for each pipe attribute to minimize the LCC of each pipe. The replacement schedules of individual pipelines are then determined with the aim of smoothing future investment. We achieved this by solving a scheduling problem under constraints that each pipeline is replaced over a specific time period in the vicinity of the predetermined base year. Investment smoothing is necessary to meet budget constraints. Our simulation showed that our new schedule effectively reduced and smoothed future investment.

to

Accounting for Seismic Risk in LCC-Based Optimum Pipe Replacement Planning of Water Distribution Systems

Kohei Hasegawa

Presenter:

Authors:

Kohei Hasegawa, Yasuhiro Arai and Akira Koizumi

Replacing deteriorating pipelines under limited budgets is one of the biggest water companies' concerns in the 21st century. Pipelines, which account for most water assets, incur costs through their lifecycle: installation, losses due to pipe failure, removal, etc. Therefore, risk-informed asset management based on a longitudinal perspective is necessary and pipe replacement planning models that minimize life-cycle costs (LCC) have been proposed. Furthermore, the computational time has been substantially reduced by integer-coding and local search algorithm. Seismic risk is also a key LCC element in earthquake-prone countries such as Japan. Recent research has enabled assessment of earthquake damage in water distribution systems (WDS); however, the seismic risk has not been sufficiently integrated into LCC models. For this reason, current LCC models may underestimate the importance of earlier replacement (or upgrade) to mitigate seismic risk. This study proposes a way to monetize longitudinal seismic risk in LCC of WDS, based on the earthquake catalog being developed by Japan's government. Model application to branched WDS in city M showed five to fifteen years earlier pipe replacement by taking seismic risk into account and demonstrated significant seismic risk reduction by replacing cast iron pipes (CIPs). Those application results suggest that exclusion of seismic risk may underestimate the importance of earlier replacement, especially in earthquake-prone countries, and that immediately replacing CIPs may yield substantial LCC reduction as well as seismic risk.

to

Transit flow for water distribution system topology disclosure

Mohamad Zeidan

Presenter:

Authors:

Mohamad Zeidan and Avraham Ostfeld

Water distribution systems present a significant challenge for structural monitoring. They comprise a complex network of pipelines buried underground that are relatively inaccessible. Water companies in some cases inherit water distribution systems that were poorly operated and managed by another company or municipality. Such systems usually suffer from improper planning and missing records, including proper records about the system layout (pipes lengths, locations, and diameters), said missing information is vital in managing and operating water distribution systems. In this study a new mapping method is proposed, that could assess missing information regarding the pipe system and thus providing more accurate network layout. The suggested method relies on the wave characteristic method (WCM) proposed by Wood in 1966, that describes the water hammer phenomenon as pressure waves generated at any point in a flow system where a disturbance that results in a change in flow rate is introduced. The methodology is demonstrated on several water distribution systems example applications.

to

Risk and Reliability of Water Distribution Systems under Changing Climate

Ehsan Roshani

Presenter:

Authors:

Ehsan Roshani, Yehuda Kleiner and Andrew Colombo

As climate change has the potential to impact component failure probability, it is expected to have a corresponding impact on the reliability of WDS. While the actual impact of climate change on component failure is still not fully understood and its quantification is a work-in-progress, the framework described here provides an efficient way to estimate its possible operational impact on WDSs. Historical water main break records were correlated to available freeze and thaw cycles, and future range of water main break rates were estimated using downscaled regional climate data. This updated break rates were applied to a real world water distribution system and its impacts on the network hydraulic variables were captured using a Monte Carlo simulations to account for its uncertainties. The aggregated results from the Monte Carlo simulation were compared with status que (i.e., no climate change impact), and the number of people affected by the changing climate in the system were estimated. Preliminary results indicated that the proposed approach is capable of capturing pipes that are vulnerable to the changing climate, and it can directly measure the number of people affected by it in a distribution network. Additionally, the use of parallel processing allows the proposed model to be successfully applied to large systems in a short period of time.

to