Grid4C provides Software as a Service (SaaS) analytics offerings aimed at all parties participating in the energy value chain: utilities, traders, distributers, grid operators, renewable energy power stations, electricity retailers and energy consumers.
The Grid4C portfolio consists of the following applications:
Real-time Machine-learning Big-Data engine that provides accurate load forecasts at the meter / appliance and sub-hour levels. Bottom up aggregations and top down disaggregations enable to forecast at any level of granularity (transformer / appliance / meter / customer / industry / territory level etc.)
A fully automated self-learning product that detects early warnings of meter malfunctioning patterns, caused by defects or configuration problems.
Advanced Machine Learning based product that monitors each meter separately, identifies anomalies, and takes into account customer and spatial parameters in order to detect and predict sophisticated thefts
An adaptive automated engine that provides accurate solar power forecasting based on proprietary Machine Learning algorithms.
Takes into account customer parameters, consumption behavior profiles and patterns, together with spatial data, in order to build both bottom-up and top down customer segmentations for pricing and energy efficiency programs. Clusters customers according to their usage patterns, predicts customers' response to marketing offerings
Uses advanced proprietary algorithms to disaggregate a household's total consumption in order to identify relevant appliances for customer targeting purposes. Takes into account customer parameters, consumption behavior profiles, usage patterns, and spatial data, as well as the customer's existing plans and products, in order to suggest the next best offering for a customer.
Takes into account customer parameters, consumption behavior profiles and patterns, customer complaints and spatial data, in order to predict each customer's churn probability and provide customer churn root-cause analysis.
Proprietary Machine Learning algorithms Learn the usage patterns of each customer, detect usage deviations, and automatically disaggregate irregularities to identify the appliances causing them. This automatically produces personalized recommendation messages, informing the customer about the irregularities detected, the appliance involved, and the effect on the monthly bill. Advanced visual tools are used to present a customer's personalized usage profile and usage patterns, billing forecasts and more.