Knowledge Reuse Method to Improve the Learning of Interference-Preventive Allocation Policies in Multi-Car Elevators

Alex VALDIVIELSO CHIAN  Toshiyuki MIYAMOTO 

Publication
IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences  Vol.E95-A  No.5  pp.990-995
Publication Date: 2012/05/01
Online ISSN: 1745-1337
Print ISSN: 0916-8508
Type of Manuscript: LETTER
Category: Concurrent Systems
Keyword: 
knowledge reuse,  option-based learning,  interference prevention,  multi-car elevators, 

Full Text: PDF(86.2KB)

Summary: 
In this letter, we introduce a knowledge reuse method to improve the performance of a learning algorithm developed to prevent interference in multi-car elevators. This method enables the algorithm to use its previously acquired experience in new learning processes. The simulation results confirm the improvement achieved in the algorithm's performance.