Elevator Bank Control Using Neural Network Destination Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing elevator scheduling systems fail to accurately predict the arrival times of future passengers, leading to inefficiencies in scheduling and increased average waiting times for passengers.
Innovation Solution
A system and method that utilize a neural network to predict the destination and arrival time of future passengers by analyzing their partial trajectories, formulated as an extended destination prediction problem, and outputting a multinomial distribution for destination and time probabilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If statistical estimation algorithms are used to predict future passenger arrivals, then scheduling can be performed with forecasted demand, but the estimation is laborious and often does not provide accurate representation of actual future requests
Solution Approach 1:
The patent replaces traditional statistical estimation algorithms with a machine learning-based prediction system that processes historical trajectory data, sensor data, and building information to generate accurate forecasts of future passenger arrivals, thereby improving prediction accuracy while automating the complex estimation process
Solution Approach 2:
The patent introduces an intermediate prediction layer that processes raw data from multiple sources (trajectories, sensors, building information) and generates refined forecasts of passenger arrivals before feeding them to the scheduling system, acting as a mediator between data collection and scheduling decisions
2Loss of time
If existing schedulers service only known requests and destinations, then the scheduling process is simple, but they ignore future arrivals of passengers altogether leading to increased average waiting times
Solution Approach 1:
The patent performs preliminary prediction of future passenger arrivals and destinations before the actual requests are made, allowing the scheduling system to proactively prepare elevator assignments and reduce waiting times by anticipating demand patterns in advance
Solution Approach 2:
The patent implements a dynamic scheduling system that continuously updates predictions and reoptimizes elevator assignments based on real-time passenger movements and forecasted arrivals, allowing the system to adapt to changing conditions and minimize waiting times dynamically
Data Source
AI summary
A control system for controlling motion of elevators of a bank of elevators uses a neural network trained for an extended destination prediction of a person based on a partial trajectory of the person to produce a multinomial of the extended destination prediction. The multinomial has at least two dimensions including a first dimension of destinations of the person and a second dimension of time intervals of the person arriving at the destinations of the first dimension. The control system optimizes a schedule of the bank of elevators based on the multinomial, and further controls the bank of elevators according to the schedule.