Sharif Digital Repository

This thesis develops a novel metaheuristic algorithm –Root Search algorithm- which is inspired by behavior of tree roots. The Root Search (RS) algorithm is capable of approximating an upperbound for the gap between obtained solution by itself and the optimal solution. In order to evaluating proposed algorithm’s performance, firstly a continuous genetic algorithm is defined, secondly both algorithms are implemented on several unconstrained continuous optimization problems, and finally the results are compared. Additionally this thesis reviews natural-insured metaheuristic algorithms so that the unique aspects of the RS algorithm can be distinguished  

The multi-mode resource-constrained project scheduling problem is one of the most complex and most widely used problems in the project scheduling knowledge area and operations research science. Recently, a new type of this problem which is called multi-skilled project scheduling problem has been introduced in the literature. The aim of MSPSP is to schedule the activities regarding the precedence relationships and multi-skilled human resource constraints in such a way that the need of each activity to a subset of skills should be met by the qualified employees. In this thesis, in addition to taking the constraints of the standard problem into account, we consider a state in which the... 

Road construction and maintenance constitute a significant part of country-wide infrastructure development budget each year. This indicates the importance of planning and design of individual road alignment in reducing the development cost. This study presents a meta-heuristic algorithm based on the concept of electromagnetism, which explores and finds a good route between two given points within a three-dimensional (3-D) context. It defines the region in a continuous 3-D surface, identifies initial routes, which are directed and shaped toward better positions by the help of local and global search. The final good solution is fine-tuned to produce a better alignment. The behaviour of the... 

Capacitor setting and reconfiguration are of main means of loss reduction and voltage profile improvement in distribution system operation. Cooperation between these two strategies can effectively minimize power losses in distribution system automation. Also, from the medium and long term viewpoint, capacitor placement and reconfiguration can be used for loss reduction and reliability enhancement. In this section, the sum of losses cost, capacitor investment cost and reliability cost should be minimized. Regarding the need to solve two non-linear combinatorial optimization problems simultaneously, metaheuristic algorithms are used in this thesis. Specifically, Elitist Genetic Algorithm,... 

In this thesis a multi-mode resource constrained project scheduling problem with considering project robustness,length of critical path and total quality of project is solved. Also we tried to propose a new mathematical model for this problem and we used both renewable and non-renewable resource constrained.The function is multi-objective so we couldn’t use classic optimization methods to reach globally or locally optimum solutions,then we used a multi-criteria optimization method to solve the problem.with respect to the computational complexity of problem we proposed a multi-objective metaheuristic (DE) algorithm to get to dominated optimal solutions.in our metaheuristic method particular... 

Distribution of goods by vehicle routing with split delivery and effective usage of trucks are two important problems which are useful separately and could be applied for different kinds of depots with different trucks to satisfy customers’ demand. Combination of these two NP-hard problems can play a great role to decrease transportation cost. Complexity of this problem is so much that we cannot solve big-size problems by exact methods or branch and bound method in limited time. We introduce metaheuristics methods such as Tabu Search, Simulated Annealing and Ant Colony Optimization which have been presented before, concisely. We present various constraints in routing and distribution,... 

Location-allocation problem is known as one of the classical problems in industrial engineering. The goals of these problems are finding optimal location of facilities and assigning these facilities to service demands from a given set of point. Location-allocation decisions are widespread through all organization in the world concerning private and public businesses. Therefore, in this study, a practical mathematical model is formulated and solved.In this paper we present a new approach for a budgeted constraint capacitated location-allocation problem integrated with cellular layout approach containing cell formation and group scheduling. Main decisions are composed of formation of... 

Research and information available shows that the number of natural disasters are increasing dramatically; hence disaster management is being raised as one of the most important topics in this field. The relief logistics is the greater part of disaster management model.
In this study, in order to manage the logistics activities effectively in a response phase of an earthquake, a hybrid meta-heuristic algorithm is developed for a transportation network. Then, the proposed algorithm is tested on a small part of Tehran.
 

This thesis presents an inventory control system which is based on Meta-heuristic Algorithms.In this thesis we will optimize the input values of our tree-echelon inventory model (keeping costs,stock worth,reorder costs,...) with these Algorithms. The presented method is able to determine the optimal stock level, costs of system and number of shipments in the system. The method is basically a decision support system (DES) and it is able to provide the order scheduling parameters of goods at operational level. The aim of the current research is to demonstrate that a Meta-heuristic algorithmis capable of solving complex inventory problems and with future improvements it is able to achieve... 

In current societies, reliability is an important and fundamental factor in designing and manufacturing products and it cannot be ignored anyway; because sometimes malfunction and failure in a system can cause severe financial losses and health issues. In the field of optimizing reliability by redundancy allocation, one of the structures that has drawn many researchers attention is series-parallel structure. In the models proposed for reliability optimization in series-parallel systems, there is a restricting assumption which all components of a sub-system have to be homogeneous. In this thesis, new models are presented which enable components of a sub-system be non-homogeneous. Modeling is... 

There are several algorithms in the literature to solve the problem of finding an optimal path between two points in a region. Such path satisfies certain constraints while optimizing the objective function. There are many factors in building, or finding, such paths in the real world, whose considerations complicate the solution procedure. This led the researchers to seriously consider meta-heuristics to solve the problem. This study hybridizes several suitable and effective meta-heuristic algorithms to solve the routing problem on a given topography. In this hybrid meta-heuristic algorithm an Ant System finds a horizontal path between the two points under considerations on the topography. A... 

Multi-mode resource-constrained project scheduling problem (MRCPSP) is considered one of the most complex and most frequently applied problems in the science field of project scheduling and the science of operations research, which recently a new version of it called multi-skilled project scheduling problem (MSPSP) has emerged in research studies. The purpose of this new problem is to schedule activities with respect to precedence relationships and resource constraints of the workforce types with multiple skills, such that the requirements of any activity for a subset of skills must be met by the qualified members of the project. In this study, in addition to observing the constraints and... 

Multi core processors are becoming mainstream in the computation area, wide application of multi core processors in Engineering, Industrial purposes and scientific applications made them popular among users and Manufacturers, cloud computing is a very well known Concept which technically is a Multicore processing environment, most of Businesses, enterprises and engineering works nowadays are getting done by the help of cloud services, hence, to get the best of these multi core cloud computing systems, scheduling of the tasks on the cores and virtual machines(VMs) is necessary, which is the answer to two questions: each task must be assigned to which core(VM)? And What is the order of... 

New product development (NPD) is a crucial process in maintaining a company’s competitive position and succeeding in dynamic markets. One of contemporary trends in the global economy is mass customisation that bases on modifications of existing products instead of designing everything as a new project. Product development can be achieved in design, planning or production phases. Specifications of previous products deliver information of design, cost and time of past NPD projects that can be the basis for developing new products. Given the low chance of technical success, market change and high cost research, market demand, product development management is important to increase the quality... 

The main goal of this thesis is to reduce the gap between the theoretical and actual project scheduling problems by introducing three problems and then proposing three different exact and meta-heuristic algorithms to solve them. These problems are as follows: the Resource-Constrained Project Scheduling Problem with Minimization of Total Weighted Resource Tardiness Penalty Costs (RCPSP-TWRTPC), the Resource-Constrained Project Scheduling Problem with Discounted Cash Flows of Weighted Earliness-Tardiness Costs (RCPSP-DCWET), and the Multi-Mode Resource-Constrained Project Scheduling Problem with Minimization of Total Weighted Resource Tardiness Penalty Costs (MRCPSP-TWRTPC). Also, we develop a... 

This study deals with the heuristic solution of flexible flow shop scheduling problems with unrelated parallel machines. No preemption of jobs is allowed. As an objective function, this paper considers the minimization of total tardiness. As a first step to propose solution, two different mixed integer mathematical models are developed which tackle the problem at hand. The problem is NP-hard and thus the developed mathematical program can only solve problem instances for small sizes in a reasonable amount of time. The next step is to build heuristics which can provide feasible solutions in polynomial time for larger problem instances. Then in the study develops a constructive heuristic for... 

This research deals with the problem of scheduling independent preemptive jobs on identical parallel machines, to minimize the total completion time (makespan). The transportation of an interrupted job from a machine to another machine requires considering a constant transportation delay i.e., Pm|prmp(delay)| Cmax. This delay is called transportation delay. We propose a Mixed Integer Linear Programming formulation for the research problem in case of two machines i.e., P2|prmp(delay)| Cmax. Since the research problem is shown to be NP-hard, a particle swarm optimization (PSO) algorithm is applied to solve the problem approximately. Based on the frequency of using local search procedure, three...