Predictive Elevator Call Generation From Passenger Behavior
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing elevator systems lack an efficient method to assist passengers in automatically generating calls based on their behavior patterns, leading to unnecessary destination suggestions and calls when the passenger's intentions are unclear.
Innovation Solution
The system compares a passenger's current behavior with stored patterns to predict their destination, automatically generating a call if a match is found, and provides suggested destinations on a mobile device if no clear match exists, while also considering data protection by storing connection data either on the device or in the elevator system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the system provides destination suggestions to assist passengers, then ease of operation is improved, but device complexity increases due to additional processing and display components
Solution Approach 1:
The system automatically generates destination suggestions by analyzing passenger behavior patterns and historical data without requiring manual input from passengers. The elevator control system self-services by predicting destinations and presenting options, reducing the operational burden on passengers while managing complexity through automated processing.
Solution Approach 2:
The system implements feedback loops where passenger responses to destination suggestions are collected and used to refine future predictions. Behavioral data from passengers interacting with the suggestion system feeds back into the prediction algorithm, improving accuracy over time while managing system complexity through iterative learning.
2Productivity
If the system stores connection data in the elevator system, then productivity is improved through faster processing, but data protection concerns worsen due to centralized storage risks
Solution Approach 1:
The system segments data storage and processing across multiple components - connection data is collected from various sensors and systems, processed through distributed algorithms, and stored in segmented repositories. This segmentation reduces the impact of any single point of failure or security breach while maintaining processing efficiency.
Solution Approach 2:
The system introduces intermediary layers between data collection and storage, including anonymization processes, encryption mechanisms, and buffered processing systems. These intermediaries protect raw connection data while enabling efficient processing and storage of transformed information.
3Loss of time
If the system generates automatic destination calls based on behavior patterns, then loss of time is reduced for passengers, but measurement precision requirements increase to accurately identify passenger intentions
Solution Approach 1:
The system generates partial destination suggestions rather than requiring complete certainty before presenting options to passengers. By providing multiple ranked suggestions rather than a single definitive prediction, the system reduces time loss while managing measurement precision requirements through probabilistic rather than deterministic approaches.
Solution Approach 2:
The system performs preliminary analysis of connection data and behavior patterns in advance to pre-calculate destination suggestions before passengers need to make decisions. This preliminary processing reduces the time required at the moment of call generation while distributing measurement precision requirements across the preliminary analysis phase.
4Ease of operation
If the system displays multiple destination suggestions ranked by similarity, then ease of operation is improved, but loss of information increases due to potential passenger confusion about selection
Solution Approach 1:
The system applies local quality by providing different levels of information and interaction for different passengers based on their preferences and historical behavior. Some passengers receive detailed explanations with multiple suggestions, while others receive streamlined options, allowing the system to optimize for ease of operation while preserving intent clarity for each individual user.
Data Source
Figure 1~3
Figure 4
Figure 4a
AI summary
A destination call, as an example of an action, is automatically produced in a lift system (1) when detected current connection data correspond to a stored behavioural pattern with use of the lift system (1). A communicative connection between a sensor (4) of a sensor system comprising a multiplicity of sensors (4) and a mobile electronic device (28) belonging to a passenger (30) is detected and connection data relating to the communicative connection are recorded. The connection data are compared with data stored in a memory device (15, 38) having at least one behavioural pattern in order to discern whether the connection data correspond to a stored behavioural pattern.