In the past several years, various optimization algorithms had been implemented on the project total cost minimization problem. Lately, it was developed an application that serves as a central platform that integrates all those previous implementations. Currently, such platform allows the access and execution of each one of the other utilities as modules/plugins. Each of which can be configured to execute optimizations over different projects. Although the platform already saves the optimization results, it lacks a suitable processing mechanism that would ease the cross analysis over all the utilities. Therefore, we want to include capture and analysis of results into the platform in order to properly aggregate them in a results database. Such database could then be queried for numerous purposes being the performance evaluation, across the optimization utilities, one of the first. In this project we analyze a total of five heterogeneous optimization utilities, all coded in Java, using algorithms based on dynamic programming and global optimization. The heterogeneity poses a challenge which we overcame by establishing a common results storage language, by means of XML files.
Discussion(0)
No comments yet. Be the first to comment.