Search published articles

Showing 2 results for K-Means Clustering

Aliakbar Hasani,
Volume 28, Issue 2 (6-2017)

In this paper, a comprehensive mathematical model for designing an electric power supply chain network via considering preventive maintenance under risk of network failures is proposed. The risk of capacity disruption of the distribution network is handled via using a two-stage stochastic programming as a framework for modeling the optimization problem. An applied method of planning for the network design and power generation and transmission system via considering failures scenarios, as well as network preventive maintenance schedule, is presented. The aim of the proposed model is to minimize the expected total cost consisting of power plants set-up, power generation and the maintenance activities. The proposed mathematical model is solved by an efficient new accelerated Benders decomposition algorithm. The proposed accelerated Benders decomposition algorithm uses an efficient acceleration mechanism based on the priority method which uses a heuristic algorithm to efficiently cope with computational complexities. A large number of considered scenarios are reduced via using a k-means clustering algorithm to decrease the computational effort for solving the proposed two-stage stochastic programming model. The efficiencies of the proposed model and solution algorithm are examined using data from the Tehran Regional Electric Company. The obtained results indicate that solutions of the stochastic programming are more robust than the obtained solutions provided by a deterministic model.

Ali Nadizadeh,
Volume 28, Issue 3 (9-2017)

In this paper, the fuzzy multi-depot vehicle routing problem with simultaneous pickup and delivery (FMDVRP-SPD) is investigated. The FMDVRP-SPD is the problem of allocating customers to several depots, so that the optimal set of routes is determined simultaneously to serve the pickup and the delivery demands of each customer within scattered depots. In the problem, both pickup and delivery demands of customers are fuzzy variables. The objective of FMDVRP-SPD is to minimize the total cost of a distribution system including vehicle traveling cost and vehicle fixed cost. To model the problem, a fuzzy chance-constrained programming model is proposed based on the fuzzy credibility theory. A heuristic algorithm combining K-means clustering algorithm and ant colony optimization is developed for solving the problem. To achieve an appropriate threshold value of parameters of the model, named “vehicle indexes”, and to analyze their influences on the final solution, numerical experiments are carried out.

Page 1 from 1     

© 2020 All Rights Reserved | International Journal of Industrial Engineering & Production Research

Designed & Developed by : Yektaweb