Predictive Fleet Scheduling for Autonomous Vehicle Maintenance

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current fleet-level autonomous vehicle scheduling systems lack efficient centralized management and predictive capabilities to optimize vehicle usage, safety, and maintenance, especially in large and complex fleets of flying, ground, or mixed vehicles.

Innovation Solution

A fleet management system with a fleet scheduler that provides real-time monitoring, predictive analytics, and scheduling to generate a master schedule for autonomous vehicles, incorporating vehicle health monitoring, demand prediction, and maintenance sequencing, while ensuring safety and optimizing operations through an IoT backbone and role-based access control.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If centralized fleet-level scheduling is implemented, then vehicle deployment efficiency is improved, but system complexity increases

Engineering Contradiction:
Improvevehicle deployment efficiencyVSAvoidsystem complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The fleet scheduler is divided into distinct functional modules: a monitoring layer for data collection, a prediction layer for analytics, and a scheduling layer for mission assignment. This segmentation allows each module to handle specific tasks independently, improving overall system efficiency while managing complexity through modular design.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system performs preliminary actions by gathering real-world data and generating predictive models before creating the master schedule. This advance preparation enables optimized vehicle deployment decisions to be made proactively rather than reactively, enhancing productivity while maintaining manageable system complexity through structured preprocessing.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If real-time monitoring and predictive analytics are implemented, then vehicle safety and maintenance are improved, but data processing requirements increase

Engineering Contradiction:
Improvevehicle safetyVSAvoiddata processing requirements
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The patent extracts critical data elements from the broader data stream, focusing specifically on vehicle health metrics, operational status, and predictive maintenance indicators. By isolating and processing only the most relevant data points, the system enhances vehicle safety and maintenance monitoring while avoiding the burden of processing all available data.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system performs preliminary data processing by gathering and analyzing real-world conditions before generating the master schedule. This advance analytics enables proactive identification of safety concerns and maintenance needs, improving vehicle reliability while managing data processing requirements through structured, pre-planned analysis.

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If predictive models are generated based on real-world data, then scheduling accuracy is improved, but computational load increases

Engineering Contradiction:
Improvescheduling accuracyVSAvoidcomputational load
Core Design Contradiction:
Measurement precisionVSUse of energy by stationary object

Solution Approach 1:

The system applies partial action by generating predictive models only for the specific parameters needed in scheduling decisions, rather than comprehensively modeling all possible variables. This selective approach maintains scheduling accuracy for critical factors while reducing unnecessary computational load and energy consumption.

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

Predictive models are generated in advance as part of the preliminary data processing phase, before master schedule creation. This timing allows computational intensive model generation to occur when system resources are available, improving scheduling accuracy while managing computational load through strategic timing of heavy processing operations.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3839689B1Fleet scheduler
Publication Date: 2024.03.27 TEXTRON INNOVATIONS INC
  • EP3839689B1 patent drawingFigure 1~2
  • EP3839689B1 patent drawingFigure 3A~3B
  • EP3839689B1 patent drawingFigure 4

AI summary

In an embodiment, a fleet scheduler (104) includes a processor; and a non-transitory computer-readable storage medium storing a program to be executed by the processor, the program including instructions for: gathering data (401) representing real-world conditions; generating and maintaining predictive models (403) based on the gathered data; and generating a master schedule (106) for a plurality of vehicles (112) based on the gathered data and the predictive models.