The DWMP blog – Episode 7. Choosing options; stepwise or optimise?
Martin Osborne
Water industry strategic advisor, asset planner and drainage expert Winner of the 2023 WaPUG Prize for contributions to the development of urban drainage practice
If you haven’t already seen the earlier episodes in this series, I suggest that you start from Episode 1 (https://tinyurl.com/DWMP-blog).
In Episode 2 I introduced a conceptual funnel to filter from generic options to preferred options for a DWMP.?See the diagram below.
But how do we go about it in practice?
We will already have defined a future planning scenario (see Episode 6) and will have assessed the system performance in this scenario to find the level of service failures; chiefly flooded properties and environmental impact.
We will have generic options and have filtered them down to those that are applicable to the area that we are considering; based on land use, type of sewerage system etc.?These options can be of varied size and the cost of the option will depend on the size.?This gives us the first problem and a big difference between a WRMP and a DWMP.?For a WRMP the level of service failures are measured as m3/d of water and the options are also measured as m3/d of additional water resource or reduced water use.?It is easy to see what options will be effective in meeting the targets.
For a DWMP the service failures are measured as number of properties or length of watercourse and the options are measured as flow rate, storage volume or area of impermeable surface.?Assessing options therefore requires modelling to show the reduction in service failures that each will bring.?(One exception is the provision of property level flood protection where one set of measures installed equals one property protected.)
We will require a combination of options to address each service failure and they will interact with options chosen to address other service failures elsewhere in the catchment.?Our aim is either to select the cheapest set of options to meet a required level of service or to select all of the options where benefit outweighs cost.?To make it more complicated, the benefit that an option delivers may depend on which other options are also selected.?An option on its own might not bring any benefit, but as an add on to another option it might.
How do we select the best combination of options for the catchment?
Optimise
One potential approach is to try all combinations of different options and different sizes of options and assess each combination to find the one that best meets our objectives.?This takes a lot of computer power, and if done manually also takes a lot of thinking time.?This approach can be automated using an optimiser which automatically generates potential combinations and then refines them based on the initial results.?This still takes thinking time to set it up and a lot of expensive computer power, but is easier than a manual approach.
This may give the best long-term value but has very high costs for the planning study.
领英推荐
Stepwise
An alternative approach is to rank the applicable options by which is likely to deliver best value based on previous experience.?This could be either a generic ranking common across all catchments or a ranking based on the characteristics of the specific catchment.
We then start with the option with the likely best value and find the optimum size of that option, we then add in the next ranked option and so on until we either reach our performance target or reach an option that delivers less benefit than its cost.?In the latter case, we could test the next ranked option to see if that would provide better value.
Each time we apply an option we should calculate the cost benefit that it delivers and use that to improve the ranking for the next area that we look at.?Dare I suggest that we could even share this information with other companies so that the whole industry gradually improves its methodology and the value that it delivers to customers.?
Combination
We could also potentially use a combination of the two approaches to keep planning costs down while still delivering good value.?Start with a stepwise approach to develop initial solutions and if the selected solutions are particularly complex and expensive then use this as the starting point for an optimisation analysis to refine the contribution of each of the selected options to drive improved value.?This step could be carried out later, for the detailed design to implement the plan rather than when drawing up the DWMP.
Summary
Selecting and assessing options is a lot more complicated for DWMPs than it is for WRMPs.?We therefore need a structured approach and importantly we need to monitor the performance of that approach by calculating the cost benefit at each step.?Optimisers have a potential role in driving down the investment cost for complex and expensive catchments, but we need a simpler robust method for other catchments and as a first step.
Questions
Has a stepwise approach delivered good value?
Is an optimiser worthwhile for DWMPs for all catchments?
Hi Martin, great summary, and you're absolutely right - it is horses for courses so to speak. We've been working in this space for 15 years and 12 years with the Water Sector so there is no silver bullet. We find that optimisation works well is when we team up and collaborate with industry specialists who can scope the problem in a manageable way. We hope that we're able to present some of the work at relevant conferences if we can encourage our clients to share the work. That way we can engage a wider audience and encourage discussion which can lead to further advances. But appreciate you taking to write these great blogs!