Worker Task Assignment Using Sensor-Driven Ergonomic Scheduling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for scheduling workers in production systems are deterministic and do not account for real-time changes in worker abilities or statistical variations, leading to inefficiencies and potential health risks.
Innovation Solution
A system that uses deep learning and sensor data from various sources (video, thermal, force, audio) to characterize worker actions and assign tasks based on ergonomic, skill, and time-related factors, optimizing efficiency and safety.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual scheduling methods are used based on institutional knowledge and observation, then scheduling simplicity is maintained, but scheduling accuracy and adaptability to real-time worker changes deteriorate
Solution Approach 1:
The patent replaces manual scheduling methods with an automated machine learning system that processes sensor data, video feeds, and worker performance metrics to generate optimized schedules. This substitution transitions from mechanical/manual operations to automated computational processes, improving scheduling accuracy while accepting increased system complexity.
Solution Approach 2:
The patent introduces a machine learning model as an intermediary between raw worker data and scheduling decisions. This intermediary processes multiple data sources including sensor streams, video analysis, and performance metrics to generate optimized assignments, bridging the gap between data collection and actionable scheduling outcomes.
2Adaptability or versatility
If deterministic scheduling models are used, then computational simplicity is maintained, but adaptability to real-time worker ability changes deteriorates
Solution Approach 1:
The patent implements dynamic scheduling that continuously adapts to changing worker conditions by processing real-time sensor data and reassigning tasks based on current worker abilities, fatigue levels, and performance metrics. This dynamic approach replaces static deterministic models with flexible, continuously updating assignments.
Solution Approach 2:
The system incorporates feedback loops where worker performance data, sensor measurements, and schedule outcomes are continuously fed back into the machine learning model to refine future assignments. This feedback mechanism enables the system to learn from past performance and adapt to individual worker characteristics over time.
3Productivity
If minimal observational data is used for scheduling, then data collection simplicity is maintained, but assignment optimization quality deteriorates
Solution Approach 1:
The patent implements a multi-functional data collection system that simultaneously gathers sensor data, video feeds, performance metrics, and ergonomic information through integrated sensors and cameras. This universal data collection approach serves multiple purposes including safety monitoring, performance tracking, and schedule optimization.
Solution Approach 2:
The system performs preliminary data collection and worker characterization before scheduling decisions are made. By pre-processing sensor data, video feeds, and performance metrics to create worker profiles and ability assessments, the system prepares optimized assignment recommendations in advance of actual scheduling needs.
Data Source
AI summary
Embodiments of the present invention provide a machine and continuous data set including process data, quality data, specific actor data, and ergonomic data (among others) to create more accurate job assignments that maximize efficiency, quality and worker safety. Using the data set, tasks may be assigned to actors based on objective statistical data such as skills, task requirements, ergonomics and time availability. Assigning tasks in this way can provide unique value for manufacturers who currently conduct similar analyses using only minimal observational data.