Machine Operation Time Estimation Using Historical and Live Work Rates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Managers of worksite operation systems face challenges in effectively planning and scheduling operations, distributing mobile work machines across worksites to efficiently complete tasks, and controlling machine settings due to uncertainties in operation duration and environmental factors.
Innovation Solution
A method is introduced that involves identifying the area of operation, obtaining historical work metrics, and using a model to estimate the time required for completing the operation. This method generates both an estimated time to complete (ETC) and an estimated time remaining (ETR) by considering factors like crop type, number of machines, and live work rates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If historical rate of work metrics and machine learning models are used to estimate time to complete, then measurement precision of time estimation is improved, but device complexity increases
Solution Approach 1:
A remote server is introduced as an intermediary component that hosts the machine learning model and processes time estimation calculations. This mediator handles the computational complexity centrally, allowing client devices to query for estimates without running complex models locally, thus improving measurement precision while managing device complexity through centralized processing.
Solution Approach 2:
The patent replaces manual or simple mechanical time estimation methods with automated machine learning models that process multiple input parameters (area, crop type, number of machines, historical data) to generate accurate time estimates. This substitution of mechanical/simpler systems with intelligent automated systems resolves the contradiction by achieving high precision through computational intelligence.
2Measurement precision
If multiple input parameters and machine learning models are used for time estimation, then measurement precision is improved, but difficulty of detecting and measuring increases
Solution Approach 1:
The system automatically collects and processes required parameters such as worksite area, crop type, and number of machines without requiring manual measurement or intervention. The machine learning model self-processes these inputs to generate time estimates, reducing the operational difficulty of detecting and measuring parameters while maintaining high measurement precision.
3Productivity
If real-time monitoring and weight metrics are applied during operation, then productivity is improved, but device complexity increases
Solution Approach 1:
The patent implements feedback mechanisms where the system monitors actual operation progress and compares it against estimated time parameters. Weight metrics and live estimates provide continuous feedback to adjust productivity calculations and provide accurate remaining time estimates, improving operational efficiency through adaptive feedback loops.
Solution Approach 2:
Time estimation calculations are performed in advance using historical data and machine learning models before operations begin. This preliminary action establishes baseline expectations and allocation metrics that guide real-time monitoring, allowing the system to track productivity against pre-calculated benchmarks without requiring complex real-time analysis during operation.
Data Source
AI summary
A method comprises identifying an area of worksite at which an operation is to be performed, identifying a historical rate of work metric, obtaining a model, identifying a number of work machines that are to perform the operation, identifying a crop type corresponding to the operation to be performed, identifying an ETC metric based on the area of the worksite, the historical rate of work metric, the number of work machines, the crop type, and the model. The method further comprises identifying, once the operation is initiated, a first weight metric, a second weight metric, a live estimate metric indicative of a first time until the work will be completed in the operation, and identifying an ETR metric, indicative of a second time until the work will be completed in the operation, based on the first weight metric, the ETC metric, the second weight metric, and the live estimate metric.