Predictive maintenance for wastewater treatment plant

Predictive maintenance for wastewater treatment plant

Predictive maintenance for wastewater treatment plant

Timeline

Nov 2022 – May 2024

Timeline

Nov 2022 – May 2024

Stage

Startup within corporation

Stage

Startup within corporation

Industry

Utilities, Water management

Industry

Utilities, Water management

My impact

  • Led end-to-end research & design activities for two product versions. Turning discovery phase into functional Minimum Viable Product.


  • Facilitated x2 live discovery workshops and started Product Trio meetings, that allowed us to align vision and define scope (Continuous Discovery Habits)


  • Ran 4+ rounds of usability tests with pilot customers and subject-matter experts


  • Enabled onboarding for 4 pilot companies, that allowed us to secure full funding

My impact

  • Led end-to-end research & design activities for two product versions. Turning discovery phase into functional Minimum Viable Product.


  • Facilitated x2 live discovery workshops and started Product Trio meetings, that allowed us to align vision and define scope (Continuous Discovery Habits)


  • Ran 4+ rounds of usability tests with pilot customers and subject-matter experts


  • Enabled onboarding for 4 pilot companies, that allowed us to secure full funding

context

It’s XXI centuary and yet everyday water & sewage maintenance means running from one alarm to another.

Not knowing what tools to bring, how big the problem is or when missing spare parts will be available.

How about no more?

context

It’s XXI centuary and yet everyday water & sewage maintenance means running from one alarm to another.

Not knowing what tools to bring, how big the problem is or when missing spare parts will be available.

How about no more?

Challenges

For our clients

Water infrastructure teams are often overwhelmed by alarms: scattered systems, missing spare parts, and no clue what they’re walking into.

→ Service Engineers

For field engineers, every day is a puzzle. They juggle unpredictable routes, respond to emergency calls on the fly, and often arrive without the right tools or parts. Often without proper training or often visibility into asset history.

→ Utility Managers

When repair cycles are long and spare parts are delayed, planning is a guesswork. Aging infrastructure only adds to the challenge. Teams are shrinking, experienced staff are rare, and most junior engineers need active supervision.


For the Posydon team

The initial scope was massive. The feature tree alone had to be adjusted to vague job titles of our target group, that differed per company.

Building clarity took time especially that we started with:

  • Huge scope with unclear priorities (welcome to Minimal Viable Product 🙃)

  • Corporate constraints like branding guidelines or regulatory limitations

  • Supporting two product versions at once: algorithm & expert-led

Challenges

For our clients

Water infrastructure teams are often overwhelmed by alarms: scattered systems, missing spare parts, and no clue what they’re walking into.

→ Service Engineers

For field engineers, every day is a puzzle. They juggle unpredictable routes, respond to emergency calls on the fly, and often arrive without the right tools or parts. Often without proper training or often visibility into asset history.

→ Utility Managers

When repair cycles are long and spare parts are delayed, planning is a guesswork. Aging infrastructure only adds to the challenge. Teams are shrinking, experienced staff are rare, and most junior engineers need active supervision.


For the Posydon team

The initial scope was massive. The feature tree alone had to be adjusted to vague job titles of our target group, that differed per company.

Building clarity took time especially that we started with:

  • Huge scope with unclear priorities (welcome to Minimal Viable Product 🙃)

  • Corporate constraints like branding guidelines or regulatory limitations

  • Supporting two product versions at once: algorithm & expert-led

goal

Helping water managers spot issues before they happen

The project focused on predictive maintenance, combining sensor data (SCADA), AI algorithms, and expert knowledge to:

  • Reduce unplanned failures

  • Extend asset life

  • Cut maintenance costs

  • Lower equipment and labor costs

We ran two product versions:

  1. Algorithm-based: recommendations driven by sensor data

  2. Expert-led: engineers (Subject Matter Experts) working behind the scenes

Did I mention we wanted to squeeze top engineers knowledge into AI-based solution? 🤖

goal

Helping water managers spot issues before they happen

The project focused on predictive maintenance, combining sensor data (SCADA), AI algorithms, and expert knowledge to:

  • Reduce unplanned failures

  • Extend asset life

  • Cut maintenance costs

  • Lower equipment and labor costs

We ran two product versions:

  1. Algorithm-based: recommendations driven by sensor data

  2. Expert-led: engineers (Subject Matter Experts) working behind the scenes

Did I mention we wanted to squeeze top engineers knowledge into AI-based solution? 🤖

learnings

  1. It’s always worth to meet your team in person, especially at early project stage

  2. Deliver small finished pieces (Minimum Viable Functionalities). Fast feedback is everything

  3. Do the best job explaining that prototype is not a product, but fill it with ‘real’ data

  4. Record all the necessary meetings,
    even if they intent to last 10 min max

learnings

  1. It’s always worth to meet your team in person, especially at early project stage

  2. Deliver small finished pieces (Minimum Viable Functionalities). Fast feedback is everything

  3. Do the best job explaining that prototype is not a product, but fill it with ‘real’ data

  4. Record all the necessary meetings,
    even if they intent to last 10 min max

learnings

  1. It’s always worth to meet your team in person, especially at early project stage

  2. Deliver small finished pieces (Minimum Viable Functionalities). Fast feedback is everything

  3. Do the best job explaining that prototype is not a product, but fill it with ‘real’ data

  4. Record all the necessary meetings,
    even if they intent to last 10 min max

I can honestly say that without Zuzanna’s contributions to our product, we would not be where we are today:
ready for commercialization less than a year into development

Andreas Betz
Global Product Manager Services
Xylem Inc.

Get in touch

Local time in Gdańsk, Poland

6:10 PM

🇵🇱 🇪🇺

© 2025 Fembot Studio 👾 by Zuzanna Adamczyk. All rights reserved.

Get in touch

Local time in Gdańsk, Poland

6:10 PM

🇵🇱 🇪🇺

© 2025 Fembot Studio 👾 by Zuzanna Adamczyk. All rights reserved.

Get in touch

Local time in Gdańsk, Poland

6:10 PM

🇵🇱 🇪🇺

© 2025 Fembot Studio 👾 by Zuzanna Adamczyk. All rights reserved.

Create a free website with Framer, the website builder loved by startups, designers and agencies.