Belmiro P.M. Duarte

ISO 9000: Some statistical results for a worldwide phenomenon

Since they were created in 1987, the ISO 9000 standards provide a unique experience of voluntary adoption and certification of entities, covering at present organizations of all sizes, sectors and natures across the globe. In this article we will provide some key results that derive from a statistical analysis performed over the richness of values compiled over time for the numbers of ISO 9000 certified entities in the world and its different countries, and their evolution since January 1993 all the way to December 2001, thus providing what we believe to be a pioneering contribution in this field, aimed at providing fact-based insights into, among other, the following issues. · Which countries are leading this movement in the world, both on absolute and relative terms? · Is there a maximum limit to the possible growth of entities certified in each country and/or the world? If so, what is it? Are we approaching that limit or quite far from reaching it yet? · What are the dynamics of transition to the ISO 9000:2000 standards across the globe? · Are there any relationships between economic development indicators and the corresponding numbers of ISO 9000 certified organizations in a given country? · Is it possible to forecast evolutions for the number of entities certified in a particular country and/or in the world?

Developing a projects evaluation system based on multiple attribute value theory

This paper presents the development process of an evaluation system to help the Portuguese Public Administration to choose a portfolio of projects for financing within the scope of Measure 1.5 of the Operational Program of the Portuguese Centro Region. The theoretical tool used is multiple attribute value theory, which focuses on the prescription of decisions in non-structured multiple objective decision scenarios. Problem structuring involved defining objectives in agreement with national development program and European Community policies, and attributes to measure the achievement of projects with respect to them. The approach required the assessment of value functions for each of the attributes, validation of independence conditions of decision makers and, finally, the aggregation of single attribute value functions into an overall multiple attribute value function (OMVF). All these structuring steps were carried out based on the preferences of a panel of decision makers with wide experience in managing and selecting projects within similar programs. The system developed, supported by a computer interface, is nowadays used to measure the appropriateness of projects to regional development goals; a project is chosen for financing if its value achieves the threshold. A portfolio of four projects embracing a range of characteristics rich enough to conclude about its performance illustrates its application.

An optimization‐based approach for designing attribute acceptance sampling plans

Purpose – This purpose of this paper is to present an optimization‐based approach to support the design of attribute sampling plans for lot acceptance purposes, with the fraction of non‐conforming items being modeled by a Poisson probability distribution function.Design/methodology/approach – The paper approach stands upon the minimization of the error of the probability of acceptance equalities in the controlled points of the operating curve (OC) with respect to sample size and acceptance number. It was applied to simple and double sampling plans, including several combinations of quality levels required by the producer and the consumer. Formulation of the design of acceptance sampling plans as an optimization problem, having as a goal the minimization of the squared error at the controlled points of the OC curve, and its subsequent solution employing GAMS.Findings – The results are in strong agreement with acceptance sampling plans available in the open literature. The papers approach in some scenarios ...

A semi-infinite programming based algorithm for finding minimax optimal designs for nonlinear models

Minimax optimal experimental designs are notoriously difficult to study largely because the optimality criterion is not differentiable and there is no effective algorithm for generating them. We apply semi-infinite programming (SIP) to solve minimax design problems for nonlinear models in a systematic way using a discretization based strategy and solvers from the General Algebraic Modeling System (GAMS). Using popular models from the biological sciences, we show our approach produces minimax optimal designs that coincide with the few theoretical and numerical optimal designs in the literature. We also show our method can be readily modified to find standardized maximin optimal designs and minimax optimal designs for more complicated problems, such as when the ranges of plausible values for the model parameters are dependent and we want to find a design to minimize the maximal inefficiency of estimates for the model parameters.

An optimization‐based framework for designing acceptance sampling plans by variables for non‐conforming proportions

Purpose – This paper seeks to present an optimization‐based approach to design acceptance sampling plans by variables for controlling non‐conforming proportions in lots of items. Simple and double sampling plans with s known and unknown are addressed. Normal approximation distributions proposed by Wallis are employed to handle plans with s unknown. The approach stands on the minimization of the average sampling number (ASN) taking into account the constraints arising from the two point conditions on the operating characteristic (OC) curve. The resulting optimization problems fall under the class of mixed integer non‐linear programming (MINLP), and are solved employing GAMS. The results obtained strongly agree with classical acceptance sampling plans found in the literature, although outperforming them in some cases, and providing a general approach to address other cases.Design/methodology/approach – The approach takes the form of formulation of the design of acceptance sampling plans by variables for non...

Optimal sizing, scheduling and shift policy of the grinding section of a ceramic tile plant

This paper addresses the optimal design of the grinding section of a ceramic tile plant operating in a cyclic mode with the units (mills) following a batch sequence. The optimal design problem of this single product plant is formulated with a fixed time horizon of one week, corresponding to one cycle of production, and using a discrete-time resource task network (RTN) process representation. The size of the individual units is restricted to discrete values, and the plant operates with a set of limited resources (workforce and equipment). The goal is to determine the optimal number and size of the mills to install in the grinding section, the corresponding production schedule, and shift policy. This problem involves labor/semi-labor intensive (LI/SLI) units with a depreciation cost of the same order as that of the operation cost. The optimal design of the grinding section comprises the trade-off between these two costs. The resulting optimization formulation is of the form of a mixed integer linear programming (MILP) problem, solved using a branch and bound solver (CPLEX 9.0.2). The optimal solution is analyzed for various ceramic tile productions and different shift policies. Scope and Purpose: This paper addresses an optimal design case study of the grinding section of a ceramic tile plant with respect to the net capacity to install, the operation scheduling and the shift policy to implement. A mathematical programming model is formulated based on the resource task network (RTN) framework representation. The problem is solved using a branch and bound algorithm. The main goal of this work is to apply optimal design/scheduling general tools to real problems commonly found in the ceramic industry sector, a particular case of labor intensive plants. The application of these methodologies to this case study demonstrates as well the importance of adopting suitable optimization strategies for plant design in order to improve the economical performance of the production lines.

Optimal Design of Acceptance Sampling Plans by Variables for Nonconforming Proportions When the Standard Deviation Is Unknown

This article presents an optimization-based approach for the design of acceptance sampling plans by variables for controlling nonconforming proportions when the standard deviation is unknown. The variables are described by rigorous noncentral Student’s t-distributions. Single and double acceptance sampling (AS) plans are addressed. The optimal design results from minimizing the average sampling number (ASN), subject to conditions holding at producer’s and consumer’s required quality levels. The problem is then solved employing a nonlinear programming solver. The results obtained are in close agreement with previous sampling plans found in the literature, outperforming them regarding the feasibility.

A Semi-Infinite Programming based algorithm for determining T-optimum designs for model discrimination

T-optimum designs for model discrimination are notoriously difficult to find because of the computational difficulty involved in solving an optimization problem that involves two layers of optimization. Only a handful of analytical T-optimal designs are available for the simplest problems; the rest in the literature are found using specialized numerical procedures for a specific problem. We propose a potentially more systematic and general way for finding T-optimal designs using a Semi-Infinite Programming (SIP) approach. The strategy requires that we first reformulate the original minimax or maximin optimization problem into an equivalent semi-infinite program and solve it using an exchange-based method where lower and upper bounds produced by solving the outer and the inner programs, are iterated to convergence. A global Nonlinear Programming (NLP) solver is used to handle the subproblems, thus finding the optimal design and the least favorable parametric configuration that minimizes the residual sum of squares from the alternative or test models. We also use a nonlinear program to check the global optimality of the SIP-generated design and automate the construction of globally optimal designs. The algorithm is successfully used to produce results that coincide with several T-optimal designs reported in the literature for various types of model discrimination problems with normally distributed errors. However, our method is more general, merely requiring that the parameters of the model be estimated by a numerical optimization.

The expected utility theory applied to an industrial decision problem: what technological alternative to implement to treat industrial solid residuals

Abstract This paper presents a method developed to solve industrial decision problems based on Expected Utility Theory. The approach involves five steps: (1) Structuring of the decision problem — this task encompasses the listing of the main objectives, technological alternatives available and the definition of the attributes of the decision problem. (2) construction of the scales to measure the attributes. (3) testing of the independence conditions needed to validate the final multiattribute utility model. (4) assessment of utility trade-offs between the attributes and construction of the final multiattribute utility model. (5) assessment of the utility of every alternative with respect to all the attributes and subsequent ranking of them. This method has been used to choose the best technological solution to treat industrial solid residuals produced by a particular mill. In order to apply the method information was gathered by interviewing the technical staff. Finally, the sensitivity of the solution to the weights of the most important attributes was evaluated. The Multiattribute Utility Theory proved to be a very suitable approach to deal with industrial decision problems. Furthermore, the sensitivity analysis corroborate the conflicts of the decision panel regarding the criteria involved. Scope and purpose The purpose of this paper is to present an approach based on the Expected Utility Theory to solve an industrial decision problem — the choice of the best technical alternative to deal with industrial solid residuals. The decision model is built based on choices elicited from the technical staff of a particular mill. The sensitivity of the solution with respect to the decision weights is evaluated in order to establish the regions of dominance of each technical solution. The best solution in present economical and strategic scenario is to burn the residuals into a fluidised-incinerator bed in the mill.

Adaptive grid semidefinite programming for finding optimal designs

We find optimal designs for linear models using a novel algorithm that iteratively combines a semidefinite programming (SDP) approach with adaptive grid techniques. The proposed algorithm is also adapted to find locally optimal designs for nonlinear models. The search space is first discretized, and SDP is applied to find the optimal design based on the initial grid. The points in the next grid set are points that maximize the dispersion function of the SDP-generated optimal design using nonlinear programming. The procedure is repeated until a user-specified stopping rule is reached. The proposed algorithm is broadly applicable, and we demonstrate its flexibility using (i) models with one or more variables and (ii) differentiable design criteria, such as A-, D-optimality, and non-differentiable criterion like E-optimality, including the mathematically more challenging case when the minimum eigenvalue of the information matrix of the optimal design has geometric multiplicity larger than 1. Our algorithm is computationally efficient because it is based on mathematical programming tools and so optimality is assured at each stage; it also exploits the convexity of the problems whenever possible. Using several linear and nonlinear models with one or more factors, we show the proposed algorithm can efficiently find optimal designs.