Imagine this, your team is operating a wastewater treatment plant. Everything is going as expected until… Oh no! A pump failed and the primary treatment train is offline.

What does your team do now? Add more chlorine? Increase other treatment trains? Cross their fingers and hope for the best? Thankfully the senior operator is on duty and he has seen this happen before. He directs the team to slightly increase chlorine dosing and increase the other treatment train pumps by 15% and that should make it for the next few days.

What happens when that senior operator is gone? Not just for the weekend but is now living it up in Margaritaville on a permanent vacation after his years of service. How can you capture his experience and knowledge now before that retirement comes?

Because of this scenario, many utilities are looking at how to incorporate the latest technology wave, AI. Imagine having the knowledge of a senior operator on duty 24/7. Someone who knows what to do in every emergency, what equipment is reliable, and keeps a perfect log of chemical dosing and effluent quality.

 There are self-made and commercially available AI software out there today that connect with SCADA, work order systems, lab data, and more to give you a virtual operator who is consistent and informed. Some are running plants almost on auto-pilot but more often they are just always in the background with recommended dosing and operations advice for the crew on shift.

 Sounds great right? It can be…but there is a big problem for most utilities…Data

 This is because what AI needs more than anything else is data…good consistent data. Most treatment plant facilities have on-site SCADA that has been collecting data for years and years so it’s a good place to start. 

What type of data do you need? It depends…quite a bit actually but below is a good starting point.

 Plant AI Assistant Data Starter List

All that sounds great right? Your SCADA likely has all this. Not so fast, in additional to SCADA you need other data. Some is easy to backfill such as weather. Other data is very tedious to backfill. Data like blower No. 7 was offline for 5 months for repairs but you did a quick swap to a unit with half the Hp, but SCADA didn’t know the difference. Or when one of your staff determined it was best to open the manual bypass by 10% to keep the equipment running and prevent anything from seizing. That may have been a good decision, but your SCADA didn’t track or log that.  This takes some of the most important data and places it outside the AI ability to read and use.

So, what should you be doing now to be ready?

• Make sure your SCADA is storing all the data, typically in the cloud or offsite in a SCADA historian.

• Make sure your work order software or at least somewhere document manual changes to equipment, gates, etc.

• For those emergency events, clearly capture what was done and what were the results. This can be manually added to the database for AI to use later.

What if you are ready now?

• Are you looking for just better automation and equipment efficiency?  Or on the job training and 24/7 monitoring?

• What does success look like? Faster decisions? Increased permit compliance?

• Who is responsible for the AI software? Operations or engineering? Maybe even IT?

• Will you make this in-house? Or prefer to use something more off the shelf?

Regardless of where your team is on their digital journey, AI is here to stay and hopefully help your team. Pass this onto your lead operator or engineer so they can start planning for the future as well.  

Make sense to connect?

Are you working to help your utility or technology company take the next step forward? I really enjoy connecting with others who share the same goals and passion of using technology to solve water utility challenges. If you want to talk through a challenge or share something interesting your team did please shoot me a note at [email protected] or schedule a 15 minute call here.