coagtech.com

What It Is

Coagtech is an AI/machine learning optimization solution for water treatment plants, with the core goal of helping operators improve coagulant chemical dosing decisions. It is not a general-purpose AI tool; instead, it builds site-specific custom models for individual treatment plants based on the coagulation process, historical operating data, and engineering expertise.

Core Capabilities and Use Cases

According to the website, Coagtech combines historical data, operational insights, and engineering knowledge to generate machine learning models for dosing recommendations. Its typical value propositions include reducing decision-making pressure on operators, increasing confidence in dosing decisions, improving the reliability of treated water quality, and, according to its claims, reducing coagulant usage by up to 20% while lowering the environmental impact of treatment residuals.

Its implementation process is divided into three phases. First, an on-site system assessment is conducted to identify process blind spots and data completeness issues, and an initial model is developed to evaluate feasibility. Second, site-specific models are developed using the plant’s historical data, with custom hardware integrated into the existing SCADA system. Finally, annual maintenance is performed, with the model reassessed and retrained based on changes in the process and source water to prevent model performance degradation.

Pricing, Integration, and Data Security

The website does not disclose specific pricing, trials, or free quotas, so procurement cost and delivery timelines need to be confirmed with the vendor. Based on the description, it appears more like an engineering consulting plus customized hardware/software deployment offering than a standard SaaS subscription tool.

In terms of integration, Coagtech explicitly states that it integrates with existing SCADA systems through custom hardware. Notably, this hardware is not connected to the internet, and the vendor emphasizes that it does not introduce additional cybersecurity risk to the water plant. This is important for critical infrastructure scenarios, but the webpage does not further explain details such as data encryption, access control, auditing, or compliance.

Pros, Cons, and Best-Fit Users

Its strengths are a highly focused use case, the ability to combine AI models with water treatment engineering knowledge, and consideration of data quality, process changes, and model maintenance. Local deployment without internet connectivity also better fits the security requirements of conservative industrial sites. The drawbacks are limited public information, with a lack of algorithm metrics, real-world case studies, pricing, and service SLA details. It is also highly dependent on the quality of each site’s historical data, and has weaker generalization and fast self-service adoption capabilities.

It is better suited for water treatment plants, engineering firms, or utility operations teams that have stable historical operating data and want to reduce chemical consumption while improving the stability of effluent water quality.

Access from China and Alternatives

Access from mainland China, Chinese-language support, and payment methods are not mentioned in the main website content, so actual usability is unknown. For deployment in China, it may be necessary to prioritize evaluating local SCADA compatibility, cybersecurity requirements, on-site service capability, and contract/payment arrangements. Alternative directions could include domestic water utility automation vendors, industrial AI optimization platforms, or dosing optimization systems custom-developed by water treatment engineering companies.