Igor Machado Coelho

A hybrid heuristic algorithm for the open-pit-mining operational planning problem

This paper deals with the Open-Pit-Mining Operational Planning problem with dynamic truck allocation. The objective is to optimize mineral extraction in the mines by minimizing the number of mining trucks used to meet production goals and quality requirements. According to the literature, this problem is NP-hard, so a heuristic strategy is justified. We present a hybrid algorithm that combines characteristics of two metaheuristics: Greedy Randomized Adaptive Search Procedures and General Variable Neighborhood Search. The proposed algorithm was tested using a set of real-data problems and the results were validated by running the CPLEX optimizer with the same data. This solver used a mixed integer programming model also developed in this work. The computational experiments show that the proposed algorithm is very competitive, finding near optimal solutions (with a gap of less than 1%) in most instances, demanding short computing times.

An ILS-based algorithm to solve a large-scale real heterogeneous fleet VRP with multi-trips and docking constraints

Distribution planning is crucial for most companies since goods are rarely produced and consumed at the same place. Distribution costs, in addition, can be an important component of the final cost of the products. In this paper, we study a VRP variant inspired on a real case of a large distribution company. In particular, we consider a VRP with a heterogeneous fleet of vehicles that are allowed to perform multiple trips. The problem also includes docking constraints in which some vehicles are unable to serve some particular customers, and a realistic objective function with vehicles’ fixed and distance-based costs and a cost per customer visited. We design a trajectory search heuristic called GILS-VND that combines Iterated Local Search (ILS), Greedy Randomized Adaptive Search Procedure (GRASP) and Variable Neighborhood Descent (VND) procedures. This method obtains competitive solutions and improves the company solutions leading to significant savings in transportation costs.

A multi-objective green UAV routing problem

Introduce a new time-dependent UAV heterogeneous fleet routing problem.Consider several objective functions and respect drones operational requirements.Design a MILP model in order to find sets of non-dominated solutions.Consider a model able to tackle multi-layer scenarios with package exchanging points.Integrate UAV into the new concepts of mini/microgrid systems inside smart cities. This paper introduces an Unmanned Aerial Vehicle (UAV) heterogeneous fleet routing problem, dealing with vehicles limited autonomy by considering multiple charging stations and respecting operational requirements. A green routing problem is designed for overcoming difficulties that exist as a result of limited vehicle driving range. Due to the large amount of drones emerging in the society, UAVs use and efficiency should be optimized. In particular, these kinds of vehicles have been recently used for delivering and collecting products. Here, we design a new real-time routing problem, in which different types of drones can collect and deliver packages. These aerial vehicles are able to collect more than one deliverable at the same time if it fits their maximum capacity. Inspired by a multi-criteria view of real systems, seven different objective functions are considered and sought to be minimized using a Mixed-Integer Linear Programming (MILP) model solved by a matheuristic algorithm. The latter filters the non-dominated solutions from the pool of solutions found in the branch-and-bound optimization tree, using a black-box dynamic search algorithm. A case of study, considering a bi-layer scenario, is presented in order to validate the proposal, which showed to be able to provide good quality solutions for supporting decision making.

Hybrid self-adaptive evolution strategies guided by neighborhood structures for combinatorial optimization problems

This article presents an Evolution Strategy (ES)--based algorithm, designed to self-adapt its mutation operators, guiding the search into the solution space using a Self-Adaptive Reduced Variable Neighborhood Search procedure. In view of the specific local search operators for each individual, the proposed population-based approach also fits into the context of the Memetic Algorithms. The proposed variant uses the Greedy Randomized Adaptive Search Procedure with different greedy parameters for generating its initial population, providing an interesting exploration–exploitation balance. To validate the proposal, this framework is applied to solve three different -Hard combinatorial optimization problems: an Open-Pit-Mining Operational Planning Problem with dynamic allocation of trucks, an Unrelated Parallel Machine Scheduling Problem with Setup Times, and the calibration of a hybrid fuzzy model for Short-Term Load Forecasting. Computational results point out the convergence of the proposed model and highlight its ability in combining the application of move operations from distinct neighborhood structures along the optimization. The results gathered and reported in this article represent a collective evidence of the performance of the method in challenging combinatorial optimization problems from different application domains. The proposed evolution strategy demonstrates an ability of adapting the strength of the mutation disturbance during the generations of its evolution process. The effectiveness of the proposal motivates the application of this novel evolutionary framework for solving other combinatorial optimization problems.

The Single Vehicle Routing Problem with Deliveries and Selective Pickups in a CPU-GPU Heterogeneous Environment

In this work, we propose a new algorithm to solve a variant of the Vehicle Routing Problem that is the Single Vehicle Routing Problem with Deliveries and Selective Pickups (SVRPDSP). Our algorithm produces good quality solutions that are better than the best known solutions in the literature. In order to reduce the time spent to solve large-sized instances, we also propose here a parallel implementation of our algorithm that explores a heterogeneous environment composed of a CPU and a GPU. Therefore, our algorithm harnesses the tremendous computing power of the GPU to improve the performance of the local searches computation. We obtained average speedups from 2.73 to 16.23 times with our parallel approach.

Exploring parallel multi-GPU local search strategies in a metaheuristic framework

Abstract Optimization tasks are often complex, CPU-time consuming and usually deal with finding the best (or good enough) solution among alternatives for a given problem. Parallel metaheuristics have been used in many real-world and scientific applications to efficiently solve these kind of problems. Local Search (LS) is an essential component for some metaheuristics and, very often, represents the dominant computational effort accomplished by an algorithm. Several metaheuristic approaches try to adapt traditional LS models to parallel platforms without considering the intrinsic features of the available architectures. In this work, we present a novel local search strategy, so-called Distributed Variable Neighborhood Descent (DVND), specially designed for CPU and multi-GPU environment. Furthermore, a new neighborhood search strategy, so-called Multi Improvement, is introduced, taking advantage of GPU massive parallelism in order to boost up LS procedures. A hard combinatorial problem is considered as case of study, the Minimum Latency Problem (MLP). For tackling this problem, a hybrid metaheuristic algorithm is considered, which combines good quality initial solutions, generated by a Greedy Randomized Adaptive Search Procedures, with a flexible and powerful refinement procedure, inside the scope of an Iterated Local Search. The DVND was compared to the classic local search procedures, producing results that outperformed the best known sequential algorithm found in the literature. The speedups ranged from 7.3 to 13.7, for the larger MLP instances with 500 to 1000 clients. Results demonstrate the effectiveness of the proposed techniques in terms of solution quality, performance and scalability.

A Hybrid CPU-GPU Local Search Heuristic for the Unrelated Parallel Machine Scheduling Problem

This work addresses the development of a hybrid CPU-GPU local search heuristic for the unrelated parallel machine scheduling problem. In this scheduling problem setup times are sequence-dependent and also machine-dependent. The objective is to minimize the maximum completion time of the schedule, known as make span. Since the problem belongs to the NP-hard class there is no known polynomial time algorithm to solve it, so metaheuristics and local search heuristics are usually developed to find good near optimal solutions. In general, the local search is the most expensive part of the heuristic method, so our algorithm harnesses the tremendous computing power of the GPU to decrease the local search computational time. We use the local search based on swapping jobs in different machines, since it is able find good near optimal solutions as we report from previous results in literature. We show that the hybrid CPU-GPU local search achieves average speedups from 10 to 27 times in relation to the pure CPU local search.

A Hybrid Heuristic in GPU-CPU Based on Scatter Search for the Generalized Assignment Problem

Abstract In the Generalized Assignment Problem (GAP), tasks must be allocated to machines with limited resources, in order to minimize processing costs. This problem has several industrial applications and often appears as a substructure in other combinatorial optimization problems. We propose a hybrid method inspired by Scatter Search metaheuristic, that efficiently generates a pool of solutions using a Tabu list criteria and an Ejection Chain mechanism. Common characteristics are extracted from the pool and solutions are combined by exploring a restricted search space, as a Binary Programming (BP) model. This method was implemented as a parallel approach to run in a Graphics Processing Unit (GPU). Experimental results show that the proposed method is very competitive to the algorithms found in the literature. On average, a gap of 0.09% is obtained over a set of 45 instances, when compared to lower bounds. Due to the integration of the method with an exact BP solver, it was capable of proving the optimality of small size instances, also finding new best known solutions for 21 instances. In terms of computational times, the proposed method performs on average 8 times faster than literature, also indicating that the proposed approach is scalable and robust for practical applications.

Generic Pareto local search metaheuristic for optimization of targeted offers in a bi-objective direct marketing campaign

Cross-selling campaigns seek to offer the right products to the set of customers with the goal of maximizing expected profit, while, at the same time, respecting the purchasing constraints set by investors. In this context, a bi-objective version of this NP-Hard problem is approached in this paper, aiming at maximizing both the promotion campaign total profit and the risk-adjusted return, which is estimated with the reward-to-variability ratio known as Sharpe ratio. Given the combinatorial nature of the problem and the large volume of data, heuristic methods are the most common used techniques. A Greedy Randomized Neighborhood Structure is also designed, including the characteristics of a neighborhood exploration strategy together with a Greedy Randomized Constructive technique, which is embedded in a multi-objective local search metaheuristic. The latter combines the power of neighborhood exploration by using a Pareto Local Search with Variable Neighborhood Search. Sets of non-dominated solutions obtained by the proposed method are described and analyzed for a number of problem instances. HighlightsUse of profit variability measure in connection to the client response.Consider the Sharpe ratio index for calculating risk-adjusted profit in targeted offers.Introduction of a bi-objective direct marketing promotional campaign.Adaptation, description and use of a general Pareto local search.

GARP: A New Genetic Algorithm for the Unrelated Parallel Machine Scheduling Problem with Setup Times

This work addresses the Unrelated Parallel Machine Scheduling Problem where setup times are sequence-dependent and machine-dependent, the UPMSPST. The maximum completion time of the schedule, known as makespan, is considered as the objective to minimize. The UPMSPST is often found in industries and belongs to the NP-hard class. Aiming to its resolution, is proposed an algorithm named GARP. This algorithm is based on Genetic Algorithm (GA) combined with Variable Neighborhood Descent (VND) and Path Relinking (PR). In addition, is used a local search method based on a Mixed-Integer Programming (MIP) model to solve the Asymmetric Traveling Salesman Problem (ATSP). The developed algorithm explores the solution space using multiple insertions and swaps movements. GARP was tested using benchmark instances and the computational results showed that it is able to produce better solutions than the algorithms found in literature, with lower variability and setting new upper bounds for the majority of the test problems.