Bin-Level Waste Quantification via Sensor Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Individuals and households lack specific metrics to inform them about their waste generation, making it difficult to reduce waste effectively, as existing waste tracking systems provide only high-level metrics that are disconnected from individual actions and their environmental impact.
Innovation Solution
A system that continuously captures data from sensors associated with waste receptacles, processes this data to determine waste properties, and provides feedback and recommendations to users through a local device and remote server, using machine learning and AI to classify waste and offer real-time notifications on proper disposal methods.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If high-level waste metrics are published quarterly or yearly for a service area, then waste monitoring coverage is achieved, but the information is insufficient to inform individuals about their specific waste generation and impact
Solution Approach 1:
The patent segments waste monitoring from aggregate service-area level to individual household/bin level. Sensors are placed in specific receptacles to measure waste independently, and data is segmented by household, bin type, and waste category. This segmentation enables precise individual feedback while maintaining overall service area monitoring capability.
Solution Approach 2:
The system implements feedback loops where waste measurement data is processed and returned to households in real-time or near-real-time. Users receive notifications about their waste generation, comparisons to targets, and educational content. This feedback transforms passive aggregate statistics into active individual information that can drive behavior change.
2Device complexity
If waste tracking systems provide only high-level metrics, then system complexity is reduced, but the disconnect from individual actions and environmental impact remains
Solution Approach 1:
The patent introduces an intermediary processing layer between simple sensor measurements and user-facing information. AI/ML models analyze sensor data to classify waste types, estimate volumes, and generate meaningful metrics. This intermediary layer translates raw sensor data into actionable insights without requiring complex user-side processing, maintaining system manageability while delivering rich information.
Solution Approach 2:
The system enables households to automatically track and analyze their own waste generation without requiring manual input or intervention. Sensors continuously monitor receptacles, and the system automatically processes data, compares it to targets, and provides feedback. This self-service approach reduces operational complexity while delivering personalized waste management information.
3Measurement precision
If sensors continuously capture data from waste receptacles, then waste measurement precision is improved, but device complexity and data processing requirements increase
Solution Approach 1:
The system implements continuous monitoring capability but processes data at appropriate intervals rather than truly continuously. Sensors capture data continuously, but processing occurs based on triggers such as weight changes exceeding thresholds, time-based sampling, or event-driven updates. This partial action approach maintains measurement precision while reducing unnecessary processing complexity and energy consumption.
Solution Approach 2:
The patent replaces manual waste tracking and classification with automated sensor systems and AI/ML processing. Instead of mechanical or manual measurement methods, electronic sensors continuously capture weight, volume, and composition data. This substitution enables precise continuous measurement without proportionally increasing operational complexity, as the automated system handles data collection and analysis.
4Measurement precision
If AI and machine learning models are used to classify waste, then waste categorization accuracy is improved, but computational requirements and processing time increase
Solution Approach 1:
The system performs preliminary data processing and feature extraction before applying complex AI/ML classification models. Sensors capture raw data, which is pre-processed to extract relevant features such as weight trends, volume estimates, and material characteristics. This preliminary action reduces the computational burden on classification models and enables faster processing while maintaining high accuracy.
Solution Approach 2:
The system implements periodic classification updates rather than continuous real-time classification for all data points. AI/ML models are applied at strategic intervals such as when receptacles are filled, at scheduled times, or when significant changes occur. This periodic action maintains classification accuracy for decision-making while reducing overall computational requirements and processing time compared to continuous classification of every data point.
Data Source
AI summary
A method includes receiving data from a device, the data includes at least a first amount of a first type of waste. The method includes comparing the data with a profile including historical waste data from a location, and sending notifications to the device based on the comparing operation. The notifications include information relating to the data and the profile. The device communicates the notifications to a user. The method may include receiving first goals from the device, and determining second goals based on community waste averages. The profile may include the first goals and the second goals. The method may include sending recommendations to the device based at least in part on the data and the profile. An exemplary device includes at least one sensor associated with a waste receptacle, which may be a computer vision imaging sensor.