The Problem
IceCOLD® distributes through a network of dealers and installation technicians across US and Asia markets. Managing the end-to-end customer onboarding journey — from site inventorization and value estimation through proposal generation, performance management, and analytics — required a structured, data-driven approach. Without a centralised platform to preprocess HVAC asset data and correlate pre- and post-installation performance, quantifying efficiency gains and scaling deployment across multiple customers and geographies was operationally complex.
The Solution
The AI Data Preprocessor was deployed to automate and structure the IceCOLD® customer onboarding and performance measurement journey across US and Asia markets. The AnySource Data Combiner ingested HVAC asset data, energy consumption records, and environmental signals, while the Data Context Builder enriched them with operational context, weather correlations, and computed efficiency metrics. This preprocessed output fed into NextqoreAI Analytics — enabling five operational workflows: inventorization, customer value estimation, proposal generation, performance management, and analytics.
AnySource Data Combiner
Energy meter data, HVAC equipment specifications (make, model, size, age, compressor and refrigerant type), GPS-tagged site images and pre- and post-installation energy bills were ingested as field device and document sources. Data was validated, normalised and structured into a unified pipeline — eliminating inconsistencies across customer locations, equipment makes and geographic markets.
Data Context Builder
The combined HVAC dataset was enriched with:
- Conditions — weather and heat index data correlated with energy consumption per run-hour across customer locations
- Correlation — equipment age, tonnage, and make cross-referenced with efficiency gain data to surface performance patterns by HVAC type
- Computation — derived efficiency indices, savings percentages, and ROI metrics calculated per unit and per customer
- OBS — Ontology Based Semantics establishing relationships between HVAC assets, installation records, energy performance, and IceCOLD® intervention timelines
The Result
- Automated inventorization — GPS-tagged asset records processed into structured HVAC portfolios with cumulative IceCOLD® requirement and handling cost estimates per customer location
- AI-driven value estimation — indicative efficiency and ROI projections delivered per customer, prioritising units with the highest savings potential before deployment
- Automated proposal generation — preprocessed asset and commercial data used to generate validated proposals, significantly reducing turnaround time
- Performance management — energy bill comparisons before and after IceCOLD® deployment tracked per unit, confirming 12–18% energy reduction criteria across installations
- Analytics dashboards — savings curves, pre- and post-intervention performance, and consumption ratios visualised across US and Asia markets — enabling IceCOLD® to demonstrate measurable efficiency gains to customers and distributors
.png)
.png)