Mehmet Emin Aydin

Coordinating metaheuristic agents with swarm intelligence

Coordination of multi agent systems remains as a problem since there is no prominent method suggests any universal solution. Metaheuristic agents are specific implementations of multi-agent systems, which imposes working together to solve optimisation problems using metaheuristic algorithms. An idea for coordinating metaheuristic agents borrowed from swarm intelligence is introduced in this paper. This swarm intelligence-based coordination framework has been implemented as swarms of simulated annealing agents collaborated with particle swarm optimization for multidimensional knapsack problem. A comparative performance analysis is also reported highlighting that the implementation has produced much better results than the previous works.

UMTS base station location planning: a mathematical model and heuristic optimisation algorithms

Radio networks of universal mobile telecommunication system (UMTS) need accurate planning and optimisation, and many factors not seen in second generation (2G) networks must be considered. However, planning and optimisation of UMTS radio networks are often carried out with static simulations, for efficiency and to save time. To obtain a good trade-off between accuracy and computational load, link-level performance factors need to be taken into account. The authors propose a mathematical model for UMTS radio network planning taking into consideration fast power control, soft handover and pilot signal power in both uplink and downlink. Optimisation strategies are investigated based on three meta-heuristics: genetic algorithm, simulated annealing (SA) and evolutionary-SA. The base station location problem is modelled as a simplified p-median problem, and parameter tuning of these meta-heuristics are presented. Extensive experimental results are used to compare the performance of different algorithms in terms of statistical measurements.

A new orthogonal array based crossover, with analysis of gene interactions, for evolutionary algorithms and its application to car door design

Recent research shows that orthogonal array based crossovers outperform standard and existing crossovers in evolutionary algorithms in solving parametrical problems with high dimensions and multi-optima. However, those crossovers employed so far, ignore the consideration of interactions between genes. In this paper, we propose a method to improve the existing orthogonal array based crossovers by integrating information of interactions between genes. It is empirically shown that the proposed orthogonal array based crossover outperforms significantly both the existing orthogonal array based crossovers and standard crossovers on solving parametrical benchmark functions that interactions exist between variables. To further compare the proposed orthogonal array based crossover with the existing crossovers in evolutionary algorithms, a validation test based on car door design is used in which the effectiveness of the proposed orthogonal array based crossover is studied.

Metaheuristic Agent Teams for Job Shop Scheduling Problems

This paper addresses and introduces an overview on various multi-agent architectures applied to teams of metaheuristic agents for job shop scheduling applications, whose developed and examined on distributed problem solving environments. We reported a couple of topologies; ATEAM is a centrally coordinating method, which provides very good results when well-studied, on the other hand, architectures based on peer-to-peer technology provide wider flexibility in implementing various fashions. The experimentation for each targeted topology has revealed more details and attracts more attentions.

A quantitative approach for measuring process innovation: a case study in a manufacturing company

Process management and innovation arguably remain among the concepts under focus of recent researches since there is no significantly outstanding method to measure and monitor the level of innovation in the manufacturing processes over a particular time period taking the fundamental activities of manufacturing processes into account. Although there are various studies relevant to process improvement, manufacturing processes are not focused on in the literature. This paper presents a novel performance indicator, called degree of process innovation, for monitoring and measuring innovation in manufacturing processes based on the four most important components among the fundamental activities of a manufacturing system. The components are namely Average Labour Utilisation, Cumulative Bottleneck Ratio, Unit Production Time and Unit Production Cost. The idea behind this approach has flourished on the basis of an indicator proposed in the literature to measure the general organisational improvements. The scope of that indicator has been narrowed down to manufacturing processes to accurately reflect the state of the manufacturing processes. The proposed approach has been verified with a case study in manufacturing industry, where each of the four sub-indicators was calculated based on the data provided and aggregated into the degree of process innovation. The innovation degree is successfully indicated.

Multiuser Scheduling on the LTE Downlink with Simulated Annealing

In this paper, the issue of multi-user radio resource scheduling on the downlink of a Long Term Evolution (LTE) cellular communication system is addressed. An optimization model has been proposed earlier, where radio resources for multiple users are jointly allocated at the air-interface. It has been shown that an optimal solution to such a problem may provide reasonable gain over a simply greedy approach. However, the complexity of such an optimal approach could be prohibitively high. By exploiting meta-heuristic method such as Simulated Annealing (SA), the results in this paper show that significant reduction in complexity can be obtained while achieving near-optimal solutions.

Multiuser scheduling in high speed downlink packet access

Multiuser scheduling is an important aspect in the performance optimisation of a wireless network as it allows multiple users to efficiently access a shared channel by exploiting multiuser diversity. For example, the 3GPP cellular standard supports multiuser scheduling in the high speed downlink packet access (HSDPA) feature. To perform efficient scheduling, channel state information (CSI) for users is required, and is obtained via their respective feedback channels. Multiuser scheduling is studied assuming the availability of perfect CSI, which would require a high bandwidth overhead. A more realistic imperfect CSI feedback in the form of a finite set of channel quality indicator values is assumed, as specified in the HSDPA standard. A global optimal approach and a simulated annealing (CSA) approach are used to solve the optimisation problem. Simulation results suggest that the performances of the two approaches are very close even though the complexity of the simulated annealing (SA) approach is much lower. The performance of a simple greedy approach is found to be significantly worse.

Multiuser scheduling on the LTE downlink with meta-heuristic approaches

Abstract In this paper, the issue of multi-user radio resource scheduling on the downlink of a Long Term Evolution (LTE) cellular communication system is addressed. An optimization model has been proposed earlier, where radio resources for multiple users are jointly allocated at the air-interface. It has been shown that an optimal solution to such a problem may provide reasonable gain over a simply greedy approach. However, the complexity of such an optimal approach could be prohibitively high. By exploiting meta-heuristic methods such as Genetic Algorithm (GA) and Simulated Annealing (SA), the results in this paper show that significant reduction in complexity can be obtained while achieving near-optimal solutions.

Heuristic-based neural networks for stochastic dynamic lot sizing problem

Multi-period single-item lot sizing problem under stochastic environment has been tackled by few researchers and remains in need of further studies. It is mathematically intractable due to its complex structure. In this paper, an optimum lot-sizing policy based on minimum total relevant cost under price and demand uncertainties was studied by using various artificial neural networks trained with heuristic-based learning approaches; genetic algorithm (GA) and bee algorithm (BA). These combined approaches have been examined with three domain-specific costing heuristics comprising revised silver meal (RSM), revised least unit cost (RLUC), cost benefit (CB). It is concluded that the feed-forward neural network (FF-NN) model trained with BA outperforms the other models with better prediction results. In addition, RLUC is found the best operating domain-specific heuristic to calculate the total cost incurring of the lot-sizing problem. Hence, the best paired heuristics to help decision makers are suggested as RLUC and FF-NN trained with BA.

A multi-agent based approach for change management in manufacturing enterprises

Change management becomes an unavoidable necessity for manufacturing enterprises. Since change in business processes carries significant impact on the performance of manufacturing companies, a change management model is definitely required to remain competitive. Moreover, utilizing agent based systems will provide computational provision and integrity to manage and measure the capabilities to follow the change in a progressive approach by employing the cooperation and collaboration properties of various agents helping for retrieval of the required information in a rapid way. Therefore, in this paper, a multi-agent based change management model is proposed to handle the changes in manufacturing enterprises. The model is validated through a case study done to measure the performance of change management capabilities in a manufacturing company. A sensitivity analysis on the results of this case study is also conducted to reveal the system reactivity to various parameters.