Load Shape Analysis for Targeted Energy Program Communication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Utility companies lack a practical way to inform energy customers about available energy efficiency programs and cannot effectively distinguish between customers likely to participate in such programs from those unlikely to participate, leading to wastage of resources and customer irritation due to irrelevant information.
Innovation Solution
A computer-implemented technique using supervised machine learning to train a predictive model that identifies energy customers' propensity to participate in energy efficiency programs based on load shapes, demographic data, and prior participation history, allowing targeted communication and resource conservation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If information about all energy efficiency programs is transmitted to all energy customers, then all customers receive complete information about available programs, but computational and network resources are wasted and customers are irritated by irrelevant information
Solution Approach 1:
The patent segments the customer base into distinct groups based on their energy consumption patterns, demographics, and program participation history. By dividing customers into segments with similar characteristics, the system can target communications to specific segments rather than broadcasting to all customers, thereby reducing resource waste while ensuring relevant information reaches the right audiences.
Solution Approach 2:
The system performs preliminary analysis of customer data, energy consumption patterns, and program participation history before transmitting program information. This preliminary action involves training predictive models on historical data to identify which customers are likely to participate in specific programs, allowing the system to pre-filter and target communications before they are sent, thus avoiding waste of resources on unlikely participants.
2Loss of information
If information about all energy efficiency programs is transmitted to all energy customers, then all customers receive complete information about available programs, but natural resources and money are wasted on physical transmission
Solution Approach 1:
The patent segments the customer base into distinct groups based on their energy consumption patterns, demographics, and program participation history. By dividing customers into segments with similar characteristics, the system can target communications to specific segments rather than broadcasting to all customers, thereby reducing resource waste while ensuring relevant information reaches the right audiences.
Solution Approach 2:
The system extracts and removes unlikely participants from the target audience for program communications. By using predictive models to identify customers with low probability of participation, the system extracts these individuals from the distribution list, preventing wasteful expenditure on physical materials and natural resources for customers who would not engage with the program information.
3Ease of operation
If conventional computer systems group customers by generalized data, then customers are organized into subgroups, but the system cannot distinguish likely participants from unlikely participants within each subgroup
Solution Approach 1:
The patent changes the parameters used for customer classification from simple demographic categories to complex multi-dimensional parameters including energy consumption patterns, load shapes, temporal patterns, and program participation history. By transforming these parameters and using them to train predictive models, the system achieves precise identification of likely participants within each customer segment, going far beyond simple generalized grouping.
Solution Approach 2:
The system introduces predictive analytics models as an intermediary between customer data grouping and program communication targeting. These models act as a mediator that analyzes multiple customer attributes and predicts participation propensity, bridging the gap between simple customer segmentation and precise targeting by providing probabilistic assessments of which customers within each segment are most likely to participate.
4Reliability
If blanket communication is sent to all customers, then no customer is missed who might be interested, but customer satisfaction decreases due to irrelevant information
Solution Approach 1:
The patent segments the customer base into distinct groups based on their energy consumption patterns, demographics, and program participation history. By dividing customers into segments with similar characteristics, the system can target communications to specific segments rather than broadcasting to all customers, thereby reducing resource waste while ensuring relevant information reaches the right audiences.
Solution Approach 2:
The system uses historical program participation data as feedback to continuously improve its predictive models. By analyzing which customers actually participated in programs after receiving communications, the system refines its predictions of which customers are likely to participate, creating a feedback loop that enhances targeting accuracy over time and reduces irrelevant communications to customers who would not engage.
Data Source
AI summary
Systems, methods, and other embodiments associated with predicting energy efficiency program participation are described. Embodiments of a method include obtaining load data for an energy customer, and determining an empirical load shape for the energy customer based on the obtained load data. A defined load shape that most closely matches the empirical load shape is selected for the energy customer. A data structure is generated to include the defined load shape, demographic data and/or site parcel data. It is determined that the energy customer is more likely to participate in an energy efficiency program than a different energy customer by applying a trained predictive model to the data structure. Transmission of information about the energy efficiency program is controlled by assigning a higher priority to a transmission to the energy customer than to a second transmission to the different energy customer.