Elevator Configuration Prediction for Faster Problem Resolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conveyance systems, such as elevator systems, face challenges in resolving operational issues efficiently, as current methods require extensive analysis and site visits by mechanics, often without access to proven configuration parameters.
Innovation Solution
A predictive system is introduced that receives problem descriptions and current configuration parameters, determines updated configuration parameters to address the issues, and sends these updates to user devices for implementation, along with impact analyses and reasons for the problems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a mechanic performs in-depth analysis of conveyance system problems using experience and subject matter experts, then the problem resolution accuracy is improved, but the time required increases significantly (taking several days and multiple site visits)
Solution Approach 1:
The system performs preliminary analysis by collecting and analyzing historical service records, engineering specifications, and customer complaints before the mechanic arrives at the site. The predictive system pre-determines updated configuration parameters based on this pre-processed data, so that when the mechanic visits, the problem can be resolved much faster with minimal on-site testing.
Solution Approach 2:
A predictive system acts as an intermediary between the mechanic and the conveyance system. This system receives problem descriptions and current configuration parameters, then uses machine learning models trained on historical data to determine updated configuration parameters. The predictive system provides its recommendations to the mechanic, who then implements them at the site, eliminating the need for the mechanic to perform lengthy manual analysis.
2Reliability
If multiple site visits are conducted to test different configuration parameters, then the problem resolution thoroughness is improved, but the productivity decreases due to repeated travel and testing
Solution Approach 1:
The predictive system performs the thorough analysis that would otherwise require multiple site visits before the mechanic arrives. By pre-determining the updated configuration parameters using historical data and machine learning, the system eliminates the need for repeated on-site testing, allowing the mechanic to implement the solution in a single visit.
Solution Approach 2:
The system replaces the mechanical process of trial-and-error testing with a computational approach. Instead of physically testing different configuration parameters at the site through multiple visits, the predictive system uses machine learning models to calculate the optimal parameters remotely, substituting computational analysis for physical testing.
3Reliability
If the mechanic relies on available experience and expert guidance, then the problem resolution expertise is improved, but the availability of proven configuration parameters at the site is limited
Solution Approach 1:
The predictive system serves multiple functions: it accesses and analyzes historical service records, engineering specifications, and customer complaints from a centralized database, then applies machine learning models to determine updated configuration parameters. This universal system consolidates the knowledge that would otherwise be分散 among individual mechanics and experts, making proven configuration parameters available to any mechanic anywhere through the networked system.
Solution Approach 2:
The system creates a digital copy of the conveyance system's configuration and problem data, then uses this copy to determine updated parameters remotely. Instead of requiring the mechanic to have physical access to extensive reference materials or expert knowledge, the predictive system accesses a digital repository of proven configurations and applies them to the specific problem at hand.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method of providing assisted problem resolution of a conveyance system, the method including receiving, at a predictive system, a problem description of a problem at the conveyance system and current configuration parameters of the conveyance system; at the predictive system, determining at least one set of updated configuration parameters to address the problem; and sending, from the predictive system to a user device, a response including the at least one set of updated configuration parameters.