Determining optimal combination of genetic operators for flow shop scheduling

Genetic algorithms (GAs) have gained wide research and applications in production scheduling fields, but the efficiency and effectiveness of a GA significantly depend on its parameters and operators. In contrast to the rich research on determination of optimal and adaptive parameters, little research has been done on determining optimal combination of genetic operators. Different from the traditional way by trial and error, this paper presents a novel and systematical approach based on ordinal optimisation (OO) and optimal computing budget allocation (OCBA) technique to determine optimal combination of genetic operators for flow shop scheduling problems. Simulation results show that the proposed methodology is able to determine optimal combination of genetic operators and simultaneously to provide a good solution with reasonable performance evaluation for scheduling problem.     

How to carry out assembly line–cell conversion? A discussion based on factor analysis of system performance improvements

The line–cell (or line–seru) conversion is an innovation of assembly systems that has received less attention. Its essence is dismantling an assembly conveyor line and adopting a mini-assembly unit, called seru (or cell). In this paper, we discuss how to do such line–cell conversions, especially focusing on assembly cell formation (ACF) and assembly cell loading (ACL). We perform 64 arrays of full factorial experiment analysis that incorporate three factors: work stations, product types, and product lot sizes. We construct a two-objective line–cell conversion model that minimises the total throughput time (TTPT) and the total labour hours (TLH). Three non-dominated solutions obtained from the two-objective model are used to evaluate the performance of the line–cell conversion. By investigating the experimental results of the ACF and the ACL, we summarise several managerial insights that could be used to help successful line–cell conversions.     

On a branch-and-bound approach for a Huff-like Stackelberg location problem

Modelling the location decision of two competing firms that intend to build a new facility in a planar market can be done by a Huff-like Stackelberg location problem. In a Huff-like model, the market share captured by a firm is given by a gravity model determined by distance calculations to facilities. In a Stackelberg model, the leader is the firm that locates first and takes into account the actions of the competing chain (follower) locating a new facility after the leader. The follower problem is known to be a hard global optimisation problem. The leader problem is even harder, since the leader has to decide on location given the optimal action of the follower. So far, in literature only heuristic approaches have been tested to solve the leader problem. Our research question is to solve the leader problem rigorously in the sense of having a guarantee on the reached accuracy. To answer this question, we develop a branch-and-bound approach. Essentially, the bounding is based on the zero sum concept: what is gain for one chain is loss for the other. We also discuss several ways of creating bounds for the underlying (follower) sub-problems, and show their performance for numerical cases. This work has been supported by the Ministry of Education and Science of Spain through grant SEJ2005/06273/ECON. M. Elena Sáz was supported by a junior research grant of Mansholt Graduate School (Wageningen Universiteit).     

A mixed integer optimisation model for data classification

In this work, a mixed integer linear programming (MILP) model is proposed for the multi-class data classification problem using a hyper-box representation. The latter representation is particularly suitable for capturing disjoint data regions. The objective function used is the minimisation of the total number of misclassified data samples. In order to improve the training and testing accuracy of our approach, an iterative solution procedure is developed to assign potential multiple boxes to each single class. Finally, the applicability of the proposed approach is demonstrated through a number of illustrative examples. According to the computational results obtained, the proposed optimisation-based approach is competitive in terms of prediction accuracy when compared with various standard classifiers.     

Sensitivity aware NSGA-II based Pareto front generation for the optimal sizing of analog circuits

This paper deals with multiobjective analog circuit optimization taking into consideration performance sensitivity vis-a-vis parameters' variations. It mainly considers improving computation time of the inloop optimization approaches by including sensitivity considerations in the Pareto front generation process, not as a constraint, but by involving it within the used metaheuristic evolution process. Different approaches are proposed and compared. NSGA-II metaheuristic is considered. The proposed sensitivity aware approaches are showcased via two analog circuits, namely, a second generation CMOS current conveyor and a CMOS voltage follower. We show that the proposed ideas considerably alleviate the long computation time of the process and improve the quality of the generated front, as well.     

Comment on 'Active simulation of grounded inductors with CCII+ s and grounded passive elements'

This comment shows that circuits for active simulation of grounded inductors with plus-type second-generation current-conveyors (CCII + s) and grounded passive elements can be systematically derived from Sedra-Smith gyrator realization. In addition to the available realizations, new circuit topologies are presented. Some of the new topologies can yield inductance forms which are not realizable using the available topologies. Also, a new topology for realizing pure inductance without recourse to element matching conditions is presented.     

Proposals towards improving worsted processing. Part 1: identification and theoretical evaluation of possibilities for optimising the drawing sequence

The possibility of improving the evenness of worsted slivers, rovings and yarns by the simple expedient of optimising the drawing sequence is examined theoretically. The drawing sequences commonly used by mills have generally been established by trial-and-error methods, using the knowledge that the evenness of any material can be contained or improved by the practice of doubling, and made worse by drafting, but with little or no application of scientific method. A simple mathematical model of the worsted processing sequence, which has proved to be a powerful tool for both problem solving and in research, is used to investigate if and how yarn quality might be improved by simple adjustments to the number of doublings and the draft ratios used. The outcome of these theoretical studies predicts that some improvement in the quality (evenness) of yarn can be achieved, and raises questions about the appropriateness of the doublings and the draft ratios commonly used.     

Proximal schemes for parabolic optimal control problems with sparsity promoting cost functionals

Fast first-order proximal methods that solve linear and bilinear parabolic optimal control problems with a sparsity cost functional are discussed. Weak convergence of these methods is proved and, for benchmarking purposes, the proposed inexact proximal schemes are compared to an inexact semi-smooth Newton method. Results of numerical experiments are presented to demonstrate the computational effectiveness of proximal schemes and to validate the theoretical estimates.     

Optimisation of the supercritical extraction of toxic elements in fish oil

This study aims to optimise the operating conditions for the supercritical fluid extraction (SFE) of toxic elements from fish oil. The SFE operating parameters of pressure, temperature, CO₂ flow rate and extraction time were optimised using a central composite design (CCD) of response surface methodology (RSM). High coefficients of determination (R²) (0.897–0.988) for the predicted response surface models confirmed a satisfactory adjustment of the polynomial regression models with the operation conditions. The results showed that the linear and quadratic terms of pressure and temperature were the most significant (p < 0.05) variables affecting the overall responses. The optimum conditions for the simultaneous elimination of toxic elements comprised a pressure of 61 MPa, a temperature of 39.8ºC, a CO₂ flow rate of 3.7 ml min^?1 and an extraction time of 4 h. These optimised SFE conditions were able to produce fish oil with the contents of lead, cadmium, arsenic and mercury reduced by up to 98.3%, 96.1%, 94.9% and 93.7%, respectively. The fish oil extracted under the optimised SFE operating conditions was of good quality in terms of its fatty acid constituents.     

Extension of dynamic programming models for management optimisation from single structure to multi-structures level

The aim of this article is the management optimisation (inspection, maintenance and rehabilitation (IM&R)) of a group of structures. It is supposed that the optimisation is constrained by limited available budget at the beginning of each time period during the entire time horizon thus creating an economical dependence between the decisions related to each of the structures. A Lagrangian relaxation technique is used for the extension of existing dynamic programming methods from single structure to multi-structures level. The methodology is illustrated by using a Generalised Partially Observable Markov Decision Process having a decision tree composed of a sequence of two decisions at the beginning of each time period, namely an inspection decision followed by a maintenance action decision. A numerical example concerning the optimisation of IM&R of 16 different bridges is presented.