# Component-Wise MOEA/D Implementation

Modular implementation of Multiobjective Evolutionary Algorithms based on Decomposition (MOEA/D) [Zhang and Li (2007), ] for quick assembling and testing of new algorithmic components, as well as easy replication of published MOEA/D proposals.

Felipe Campelo and Lucas Batista
Operations Research and Complex Systems Laboratory - ORCS Lab
Universidade Federal de Minas Gerais
Belo Horizonte, Brazil

Claus Aranha
Faculty of Engineering, Information and Systems
University of Tsukuba
Tsukuba, Japan

R package containing a component-based, modular implementation of the Multiobjective Evolutionary Algorithm with Decomposition (MOEA/D) framework.

The MOEA/D framework is seen as a combination of specific design decisions regarding several independent modules:

• Decomposition strategy;
• Aggregation function;
• Objective scaling strategy;
• Neighborhood assignment strategy;
• Variation Stack;
• Update strategy;
• Constraint handling method;
• Termination criteria.

This package provides several options for each module, as explained in the documentation of its main function, `MOEADr::moead()`. The input structure of this function is also explained in its documentation.

To install the current release version in your system, simply use:

``````install.packages("MOEADr")
``````

For the most up-to-date development version, install the github version using:

``````# install.packages("devtools")
``````

## Example

As a simple example, we can reproduce the original MOEA/D (Zhang and Li, 2007) and run it on a 30-variable ZDT1 function:

`````` ## 1: prepare test problem
library(smoof)
ZDT1 <- make_vectorized_smoof(prob.name  = "ZDT1",
dimensions = 30)

## 2: set input parameters
problem   <- list(name       = "ZDT1",
xmin       = rep(0, 30),
xmax       = rep(1, 30),
m          = 2)
decomp    <- list(name       = "SLD", H = 99)
neighbors <- list(name       = "lambda",
T          = 20,
delta.p    = 1)
aggfun    <- list(name       = "wt")
variation <- list(list(name  = "sbx",
etax  = 20, pc = 1),
list(name  = "polymut",
etam  = 20, pm = 0.1),
list(name  = "truncate"))
update    <- list(name       = "standard",
UseArchive = FALSE)
scaling   <- list(name       = "none")
constraint<- list(name       = "none")
stopcrit  <- list(list(name  = "maxiter",
maxiter  = 200))
showpars  <- list(show.iters = "dots",
showevery  = 10)
seed      <- NULL

## 3: run MOEA/D
out1 <- moead(problem = problem,
decomp = decomp, aggfun = aggfun, neighbors = neighbors, variation = variation,
update = update, constraint = constraint, scaling = scaling, stopcrit = stopcrit,
showpars = showpars, seed = seed)

## 3.1: For your convenience, you can also use the preset_moead() function to reproduce the above setup,
##      and only modify the desired parts:

out2 <- moead(problem = problem,
stopcrit = list(list(name = "maxiter", maxiter = 1000)),
showpars = showpars, seed = 42)

# 4: Plot output:
plot(out1\$Y[,1], out1\$Y[,2], type = "p", pch = 20)
``````

Have fun!
Felipe

# News

• Changed output class of moead() to moead (instead of moeadoutput). All related functions (plot, summary) and documentation were updated.
• Added S3 print function for moead objects

• Corrected the version for the two fixes below;
• `moead()` output objects now contain the input configuration used for running the algorithm, for easier reproducibility;
• Fixed problem in `moead()` that was compromising reproducibility when running with `irace`;

• Fixed minor error in vignette "Fine tuning MOEADr using irace";
• Fixed minor error in final Archive composition in `moead()`;

• Moved package smoof back to Suggestions;
• Updated package examples to reflect looser dependency on smoof;
• Added summary and plot functions (S3);
• Added three more Vignettes to explain different aspects of the package: basic usage; automated algorithm assembling and tuning using irace; and using a user-defined operator with the MOEADr framework;
• Added warning / user confirmation in decomposition_sld() to prevent huge population sizes due to mis-specification of a user parameter;
• Added preset_moead() to provide fast access to standard configurations from the MOEA/D literature (two "standard MOEA/D" versions and one "MOEA/D-DE");

• Moved package smoof from Suggestions to Imports
• Depends updated to R (>= 3.4.0)

• Added `NEWS.md` to track changes to the package.
• R.utils is no longer imported by the MOEADr package
• Skipped some tests on Solaris platforms, which were throwing spurious errors.

• Initial release on CRAN

# Reference manual

1.1.0 by Felipe Campelo, 2 years ago

Browse source code at https://github.com/cran/MOEADr

Authors: Felipe Campelo [aut, cre] , Lucas Batista [com] , Claus Aranha [aut]

Documentation:   PDF Manual