Decision support for water planning

Define the decision before choosing software

Are you comparing supply options, prioritising monitoring sites or choosing an operating rule? Each question needs different data and outputs. Start by naming the decision, its owner and the people affected.

A water model may represent flows and storage. A multi-criteria comparison may describe trade-offs between cost, reliability, environmental effects and other objectives. Neither automatically decides whose priorities should carry the most weight.

Build an options register

Describe each option as a complete system. For example, a reuse option includes treatment, storage, distribution and operation, not only a treatment unit. Apply the same boundary and assessment period to every option.

Separate three kinds of input

• Observed: measurements with a date, method and coverage.
• Estimated: values derived from a stated method.
• Assumed: choices made to explore a scenario.

Keep this distinction in exported tables and presentations. A polished chart should not make an assumed future demand look like a recorded observation.

Test whether the choice is robust

Change uncertain inputs within justified ranges and see whether the preferred option changes. Test plausible dry periods, operating interruptions and demand changes. If small changes reverse the ranking, show that sensitivity rather than hiding it in a single score.

A numerical weight is not a measurement of social importance. Explain how criteria and weights were agreed, who participated and which perspectives remain unrepresented.

The SMART contribution

The research combined water-resource analysis with decision-support approaches in the Lower Jordan Valley. The reports describe how case studies and different kinds of evidence were brought into a planning process. They remain useful examples of method development, with the limitations and dates of their original work.

Our water-balance calculator is a transparent learning aid. It exposes a small set of inputs and arithmetic. It is not the historic SMART decision-support software and it does not calculate an optimum management plan.

Make the decision traceable

Keep a short record of the options considered, rejected alternatives, evidence sources, unresolved uncertainties and responsibilities for follow-up. That record can be more useful to the next planning team than an unexplained final ranking.

The historical Dropedia knowledge-management record shows how the SMART programme organised related concepts. This is a preserved historical page, not a live version of the original software.