Reducing Wastewater Overflows: Optimising Capital Investment in Christchurch

Christchurch has a large and complex wastewater network serving 384,000 people. Past efforts to reduce wet weather overflows were based on traditional trial-and-error modelling, and the system upgrades stemming from those modelling efforts have not been as effective as hoped, despite costing more than $150 million. Even with those upgrades in place, there remains an overflow volume of 38,000 m³ to waterways and a further 40,000 m³ overflows from 165 manholes during a 3-year ARI (average recurrence interval) storm.

This project’s objective was to identify the most cost-effective suite of projects to prevent system overflows for three different return period storms (6-month, 1-year and 3-year ARI).

To best evaluate a broad range of alternatives in the improvement plan, Christchurch City Council (CCC) chose to use genetic algorithm (GA) optimisation, Optimatics’ Optimizer WCSTM software, and an experienced optimisation team.

Optimizer WCSTM links to the hydraulic model and life-cycle cost data. Cloud computing is used to run the model continuously and evaluate thousands of combinations of improvement alternatives against total cost and hydraulic performance, enabling the optimisation team to identify solutions that meet the planning criteria at least cost.

The structured process evaluated thousands of possible combinations of pipe upgrades, pump station upgrades, inflow and infiltration (I/I) reduction, increased treatment plant capacity, flow diversion, and new storage facilities. This would not have been possible with a traditional trial-and-error modelling approach.

Improvement alternatives were gradually added in order of complexity (starting from increasing conveyance capacity along existing alignments, then flow controls and diversion options, then storage facilities, and finally I/I reduction options). This strategy enabled the team and CCC to review optimised solutions incrementally and gain appreciation of the potential cost savings associated with various alternatives.

The team first studied existing system performance results of wastewater overflow volume and frequency in the 15-year continuous simulation of rainfall from 2000 to 2015. They then devised a plan to determine the costs to abate the overflows, differentiating between 1) manholes, 2) “Priority 1” outfalls and 3) “Priority 2” outfalls. If the estimated cost was below $500/m³ to abate overflow, the outfall was designated as Priority 1. If the estimated cost was above $500/m³ to abate, the outfall was Priority 2.

Improvement alternatives and associated unit costs were then placed into the model and Optimizer, and numerous optimisation runs were performed. The final, optimised solutions showed the location and size of pipe upgrade, storage, and I/I reduction improvements that would meet the system performance requirements at least cost.

This project achieved capital cost savings of up to 32% to achieve an aspirational target of no overflows in a 3-year annual recurrence interval storm. The optimisation process also gave CCC certainty that the recommended projects were the most cost-effective suite of projects to achieve its target.