Applied Simulation and Optimization by Miguel Mujica Mota Idalia Flores De La Mota & Daniel Guimarans Serrano

Author:Miguel Mujica Mota, Idalia Flores De La Mota & Daniel Guimarans Serrano
Language: eng
Format: epub
Publisher: Springer International Publishing, Cham

The logic of the simulation model is that if no time window constraints are defined for the first trip, it starts at midnight; otherwise it starts at time to match the first customer’s window. The next trip starts immediately after the previous one, unless its start time should be delayed to satisfy time windows of customers in the route of this trip. No times are encoded, and hence, the potential solution has no directly encoded idle time, and its trips cannot overlap. Due to the high universality of this encoding, different permutation manipulation operators can be applied in the search of the optimal solution.

A grid of landscape analysis experiments is created to compare values between different landscapes. First, comparison of different mutation operators is performed. Second, comparison between existing and proposed encodings is done. Results of comprehensive analysis experiments are described in [6]. Particularly found, that in a random walk, values of autocorrelation function are slightly lower for the replacement operator. In the up-down walk the situation is the opposite: replacement mutation has higher correlation than shift mutation, but the three artificial problems are different to the others (see Fig. 15; black dots are for replacement and green for shift mutator). Moreover, the value of the autocorrelation function in random and up-down walks is lower for the permutation encoding, which means that landscapes of this encoding should be more rugged.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.