Motomtech turned APCO’s Digital Twin from a Python research script into a live web platform for optimizing water treatment operations. The original model used hybrid modeling, decision theory, and machine learning but required technical knowledge to operate. Motomtech built an accessible web interface for plant operators at the Southeast Regional Water Treatment Plant in partnership with the Jordan Valley Water Conservancy District, APCO Inc., and the University of Utah.

The Digital Twin platform delivered a 30% reduction in decision-making time through smarter chemical dosing and water source selection. Operators also gained 3x faster data interpretation through intuitive dashboards visualizing uncertainty and system recommendations, 15% lower operational costs achieved through predictive optimization, and full conversion from a research script into an accessible web platform.

The Digital Twin platform runs on Python with a hosted web application backend, and a web-based interface built on a custom Python framework. The architecture preserved the original research model logic while making it secure, intuitive, and always-on for frontline plant operators.

Motomtech ran a four-phase engagement. Phase one was conceptualization with researchers and plant personnel to understand the modeling logic, operational challenges, and user expectations, and phase two was software prototyping with an intuitive UI that simplified the model output. Phase three was implementation as a web application, and phase four was maintenance and enhancement with feedback from SERWTP operators covering improved visualizations, expanded model flexibility, and plans for live data integration.

Philip Smith, CEO at Digital Twin, said Motomtech helped take an academic model and turn it into a solution operators can actually use every day. The new platform is fast, intuitive, and built for the real world, helping the team make better decisions, save money, and manage resources more sustainably.