Method and an apparatus for determining an allocation decision for at least one elevator

Patent metadata

Field	Value
Publication number	US-11753273-B2
Application number	US-201815964827-A
Country	US
Kind code	B2
Filing date	Apr 27, 2018
Priority date	Nov 16, 2015
Publication date	Sep 12, 2023
Grant date	Sep 12, 2023

Abstract

A method for determining an allocation decision for at least one elevator includes using an existing calls in an elevator system as a first input in a machine learning module, processing the first input with the machine learning module to provide a first output comprising a first allocation decision, using the first output as a second input in an iterative module, processing the second input with the iterative module to provide a second ouput comprising a second allocation decision, and providing the second allocation decision to an elevator control module and to an allocation decision storage for further machine learning module training.

First claim

The invention claimed is:

1. A method for determining an allocation decision for at least one elevator, the method comprising:
using existing calls in an elevator system as a first input in a machine learning module;
processing the first input with the machine learning module to provide a first output comprising a first allocation decision;
using the first output as a second input in an iterative module;
processing the second input with the iterative module to provide a second output comprising a second allocation decision, wherein the iterative module implements a genetic algorithm or an ant colony algorithm to process the second input comprising the first allocation decision; and
providing the second allocation decision to an elevator control module as an allocation command and to an allocation decision storage for further machine learning module training, such that the elevator control module controls at least one elevator car based on the allocation command,
wherein the machine learning module provides the first allocation decision based on having been trained by a certain number of previous allocation decisions provided by the iterative module.

2. The method of claim 1 , further comprising:
processing, prior to using the existing calls in the elevator system as the first input to the machine learning module, the existing calls in the elevator system with the iterative module, to provide a third allocation decision;
providing the third allocation decision from the iterative module to the elevator control module and to the allocation decision storage, wherein the allocation decision storage is provided the third allocation decision as a previous allocation decision;
using, until at least one predetermined criterion is fulfilled, the previous allocation decision to teach the machine learning module; and
using, after the at least one predetermined criterion has been fulfilled, the existing calls in the elevator system as the first input in the machine learning module.

3. The method of claim 1 , further comprising:
using predicted calls in the elevator system as the first input to the machine learning module.

4. The method of claim 1 , wherein the machine learning module comprises an artificial neural network module.

5. A non-transitory computer readable medium storing program code, which when executed by at least one processor, causes the at least one processor to perform the method of claim 1 .

6. An apparatus for determining an allocation decision for at least one elevator, the apparatus comprising:
a memory storing program instructions; and
a processor configured to execute the program instructions to
use existing calls in an elevator system as a first input in a machine learning module;
process the first input with the machine learning module to provide a first output comprising a first allocation decision;
use the first output as a second input in an iterative module;
process the second input with the iterative module to provide a second output comprising a second allocation decision, wherein the iterative module implements a genetic algorithm or an ant colony algorithm to process the second input comprising the first allocation decision; and

provide the second allocation decision to an elevator control module as an allocation command and to an allocation decision storage for further machine learning module training, such that the elevator control module controls at least one elevator car based on the allocations command,
wherein the machine learning module provides the first allocation decision based on having been trained by a certain number of previous allocation decisions provided by the iterative module.

7. The apparatus of claim 6 , wherein the processor is further configured to execute the program instructions to:
process, prior to using the existing calls in the elevator system as the first input in the machine learning module, the existing calls in the elevator system, with the iterative module, to provide a third allocation decision;
provide the third allocation decision from the iterative module to the elevator control module and to the allocation decision storage, wherein the allocation decision storage is provded the third allocation decision as a previous allocation decision;
use, until at least one predetermined criterion is fulfilled, the previous allocation decision to teach the machine learning module; and
use, after the at least one predetermined criterion has been fulfilled, existing calls in the elevator system as the first input in the machine learning module.

8. The apparatus of claim 6 , wherein the processor is further configured to execute the program instructions to:
use predicted calls in the elevator system as the first input to the machine learning module.

9. The apparatus of claim 6 , wherein the machine learning module comprises an artificial neural network module.

10. The apparatus of claim 6 , wherein the machine learning module comprises a graphics processing unit.

11. An elevator system comprising:
an elevator control module configured to control at least one elevator car; and the apparatus of claim 6 .

Assignees

• Kone Corp

Inventors

• Kitov Sergey

Classifications

• G06N3/09

  Supervised learning · CPC title

• B66B1/2408 Primary

  where the allocation of a call to an elevator car is of importance, i.e. by means of a supervisory or group controller · CPC title

• G06N3/08

  Learning methods · CPC title

• B66B2201/231

  Sequential evaluation of plurality of criteria · CPC title

• G06N20/00

  Machine learning · CPC title