Predictive Cleanliness Scoring for Vehicle-Sharing Fleets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Vehicle-sharing and ride-hailing fleets lack effective predictive cleanliness analysis systems to determine when vehicles need cleaning, leading to potential user dissatisfaction and operational inefficiencies.
Innovation Solution
A method and system using a processor to assign vehicles based on contextual information such as location, route, user history, and weather, calculating cleanliness scores, and transmitting alerts to operators to activate cameras for interior and exterior inspections when scores fall below threshold values, indicating the need for cleaning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If no predictive cleanliness analysis system is implemented, then the system complexity remains low, but vehicle cleanliness management effectiveness deteriorates
Solution Approach 1:
The system performs preliminary cleanliness analysis by evaluating contextual information (route, location, weather, user history) before the vehicle is returned, predicting potential cleanliness issues in advance and enabling proactive scheduling of cleaning tasks before the vehicle is reassigned to the next user
Solution Approach 2:
The system implements feedback loops where cleanliness scores are calculated based on contextual data, alerts are generated when thresholds are exceeded, and cleaning history is recorded and fed back into future cleanliness predictions, creating a continuous improvement cycle for fleet hygiene management
2Measurement precision
If comprehensive contextual information is collected for cleanliness analysis, then cleanliness prediction accuracy improves, but data processing requirements and system complexity increase
Solution Approach 1:
The system extracts only the most relevant contextual features from available data (route characteristics, location type, weather conditions, user cleaning history) to calculate cleanliness scores, filtering out unnecessary information to maintain processing efficiency while achieving accurate predictions
Solution Approach 2:
The system transforms diverse contextual information (qualitative route descriptions, weather conditions, user behaviors) into quantifiable parameters and cleanliness scores that can be processed algorithmically, enabling accurate predictions through standardized parameter evaluation
3Speed
If cleanliness alerts are transmitted for every low score, then cleaning responsiveness improves, but operator alert fatigue and operational inefficiency increase
Solution Approach 1:
The system applies different alerting strategies based on the specific cleanliness score thresholds and vehicle contexts, transmitting alerts selectively for scores below certain thresholds while prioritizing vehicles based on their cleanliness risk profiles, ensuring responsive cleaning without overwhelming operators with unnecessary alerts
Data Source
AI summary
A method and system are disclosed and include assigning a vehicle to a user. The method also includes obtaining contextual information associated with the vehicle-sharing request, and the contextual information includes at least one of location information associated with the vehicle-sharing request, route information associated with the vehicle-sharing request, and user information associated with a vehicle-sharing account corresponding to the user. The method also includes determining a cleanliness score based on the contextual information and determining whether the cleanliness score is less than a first threshold value. The method also includes transmitting, in response to (i) the cleanliness score being less than the first threshold value and (ii) a vehicle-sharing session corresponding to the vehicle-sharing request being complete, an alert that is configured to indicate that the vehicle needs to be cleaned.