Skip to content
Silver Lake Analytics
All work

Municipal water treatment

Scheduling water treatment plants without a scheduling system

Multiple treatment plants, rotating operators, external contractors, and a spreadsheet holding it all together. Now one task standard across the workforce, time tracked against it, and reporting the regulator can be shown.

1,089
Task templates imported
21
Contractors managed in the system
6
Feedback rounds shipped after go-live

The situation

Work across several treatment plants was coordinated by spreadsheet and phone between supervisors, operators and contractors. Recurring maintenance schedules existed largely as documents and individual knowledge, which meant the quality of the schedule depended on who was doing it that week.

Nothing was broken, exactly. But there was no single place where a supervisor could see what was happening across every plant at once, and no structural reason a recurring task would be scheduled the same way twice.

What was built

A real-time scheduling and task management application, now in daily use by the operations group.

Scheduling across multiple plants, visible to supervisors and operators at the same time rather than reconciled after the fact.

A task template library, so recurring maintenance is scheduled from a known standard instead of rebuilt from memory. 1,089 templates were imported during the build.

Contractor management, with 21 external contractors assignable to work directly in the system.

Role separation between scheduler-administrators, supervisors and operators, so the schedule cannot be quietly rewritten by anyone who opens it.

SMS notification to operators.

The part that shows it is real

Two things, neither of them a metric.

The application was migrated from an earlier hosting and database stack onto a new one mid-engagement, without interrupting the team using it. Platform migrations are where most small software engagements come apart. This one was invisible to the operators.

And after go-live, six rounds of feedback from the operations manager were specified, built and shipped. Real operational software is not finished at go-live. It is finished when the people using it stop finding friction. Most consultants are gone by round two.

The honest limit

One of the two operator groups went live first by agreement, with the second deliberately deferred rather than attempted at the same time.

Sequencing the rollout meant slower coverage in exchange for a first group that was properly supported. That was the right trade and it was made on purpose.

Recognise any of this?

Thirty minutes, no obligation. Tell me what is not working and I will tell you whether it is worth building something.

Email us about your project