David de la Fuente

A decision support system for applying failure mode and effects analysis

This study describes an alternative way of applying failure mode and effects analysis (FMEA) to a wide variety of problems. It presents a methodology based on a decision system supported by qualitative rules which provides a ranking of the risks of potential causes of production system failures. By providing an illustrative example, it highlights the advantages of this flexible system over the traditional FMEA model. Finally, a fuzzy decision model is proposed, which improves the initial decision system by introducing the element of uncertainty.

A review of machine learning in dynamic scheduling of flexible manufacturing systems

A common way of dynamically scheduling jobs in a flexible manufacturing system (FMS) is by means of dispatching rules. The problem of this method is that the performance of these rules depends on the state the system is in at each moment, and no single rule exists that is better than the rest in all the possible states that the system may be in. It would therefore be interesting to use the most appropriate dispatching rule at each moment. To achieve this goal, a scheduling approach which uses machine learning can be used. Analyzing the previous performance of the system (training examples) by means of this technique, knowledge is obtained that can be used to decide which is the most appropriate dispatching rule at each moment in time. In this paper, a review of the main machine learning-based scheduling approaches described in the literature is presented.

Application of distributed intelligence to reduce the bullwhip effect

The paper applies distributed intelligence to the bullwhip or Forrester effect, which it manages to reduce in a range of time series to which a genetic algorithm was applied. The paper is divided into four parts. The first provides an overview of the Forrester, or bullwhip, effect. The second describes the genetic algorithms in terms of being devices that provide the model with intelligence, and introduces the agent network and the general model that supports them. The third describes the software used in the model described. The final section provides numeric examples and draws a number of conclusions.

Impacto de los ingresos urgentes innecesarios sobre las estancias hospitalarias en un hospital de Asturias

Fundamento: Los ingresos innecesarios en los hospitales de agudos tienen importantes consecuencias sobre su eficiencia y organizacion. El objetivo de este estudio es identificar el grado de inadecuacion de los ingresos realizados desde un servicio de urgencias hospitalario y sus causas, asi como cuantificar las estancias inadecuadas generadas por estos ingresos. Metodo: Se evaluo la adecuacion de 622 ingresos realizados en al ano 2002 seleccionados aleatoriamente, y la del total de estancias generadas por los ingresos inadecuados y una muestra representativa de los ingresos adecuados de un hospital de segundo nivel de Asturias. El instrumento de revision fue el Appropiateness Evaluation Protocol. Se realizo un analisis descriptivo de la muestra, un analisis bivariante y un analisis de regresion logistica multivariante. Resultados: Se consideraron inadecuados 63 ingresos (10,1%). La principal causa de inadecuacion fueron los ingresos para realizar pruebas diagnosticas y/o tratamientos que podrian realizarse de forma ambulatoria. Los ingresos innecesarios generaron un 78,2% de estancias innecesarias y los necesarios un 24,8%. Incrementaron el riesgo de ingresos innecesarios la derivacion a urgencias desde consultas externas del propio hospital (OR:4,50, IC 95%: 1,59-12,76), ingresar en horario de manana (OR: 13,97, IC 95%: 1,86-104,76) o tarde (OR: 7,70, IC 95%: 1,01-58,72), ingresar en los servicios de cardiologia (OR: 3,93, IC 95%: 1,22-12,70) y neurologia (OR: 5,86, IC 95%: 1,88-18,30) disminuyo el riesgo de ingreso innecesario la experiencia de ingresos previos (OR: 0,34, IC 95%: 0,18-0,65). Conclusiones: Los ingresos innecesarios generan tres veces mas estancias inadecuadas que los necesarios. Los problemas organizativos del centro son la principal causa de ingresos inadecuados.

Genetic algorithms to optimise the time to make stock market investment

The application of Artificial Intelligence described in this article is intended to resolve the issue of speculation on the stock market. Genetic Algorithms is the technique that is used, with the article focusing on the different ways that chromosomes can be designed and on how the pertinent evaluation mechanism is established. The problem will be based on the speculation systems that are typical of Technical Analysis.

Urgencias hospitalarias y de atención primaria en Asturias: variaciones entre áreas sanitarias y evolución desde 1994 hasta 2001

Objetivo: Describir la frecuencia de utilizacion de los servicios de urgencias en las areas sanitarias de Asturias desde 1994 a 2001 y analizar su variabilidad. Metodos: Se estimo la demanda de urgencias hospitalarias y de atencion primaria en las 8 areas sanitarias de Asturias, y se analizo su evolucion y las diferencias entre areas, usando la estandarizacion indirecta y estadisticos de variabilidad. Resultados: Entre 1994 y 2001 se realizaron casi 6,5 millones de urgencias (un 43,8% en hospitales), con un crecimiento medio anual del 6,2% (un 7,8% en atencion primaria y un 5,1% en hospitales) con gran heterogeneidad entre areas. La variabilidad fue mayor en atencion primaria y disminuyo en el periodo estudiado (coeficientes de variacion: 0,38 y 0,27 para 1994 y 2001 en atencion primaria, y 0,14 y 0,11 en hospitales, respectivamente). Conclusiones: La utilizacion de las urgencias crecio en el periodo estudiado y se observa una gran variabilidad entre areas sanitarias.

Dynamic scheduling of flexible manufacturing systems using neural networks and inductive learning

Dispatching rules are usually applied dynamically to schedule jobs in flexible manufacturing systems. Despite their frequent use, one of the drawbacks that they display is that the state the manufacturing system is in dictates the level of performance of the rule. As no rule is better than all the other rules for all system states, it would be highly desirable to know which rule is the most appropriate for each given condition, and to this end this paper proposes a scheduling approach that employs inductive learning and backpropagation neural networks. Using these latter techniques, and by analysing the earlier performance of the system, “scheduling knowledge” is obtained whereby the right dispatching rule at each particular moment can be determined. A module that generates new control attributes is also designed in order to improve the “scheduling knowledge” that is obtained. Simulation results show that the proposed approach leads to significant performance improvements over existing dispatching rules.

Design of a fuzzy finite capacity queuing model based on the degree of customer satisfaction: Analysis and fuzzy optimization

Due to uncontrollable factors, the parameters in the queuing models may be uncertain and so in this paper we present the design of a fuzzy finite capacity queuing model in which the arrival pattern and the service pattern follow an exponential distribution under uncertain parameter. As for the design of the system, the criterion of optimization proposed is aimed at the optimum selection of the number of servers, with the goal of providing a high degree of satisfaction to clients when they join the system. This optimization process will be solved through Markov chains with fuzzy state. The validity of the procedure proposed to incorporate fuzzy data in this queuing system is confirmed by a fuzzy simulation experiment of the differential equations, which govern the behaviour of the model, by the Bontempis Qua. Si. III algorithm. The extension of queuing decision models to fuzzy environments enables the decision maker to obtain more informative results and wider knowledge on the behaviour of the system, since the results obtained in the fuzzy queuing model are fuzzy subsets containing the whole initial information; that is why the finite capacity queuing models with uncertain data can have a broader range of applications.

Determining warehouse number and location in Spain by cluster analysis

The aim of the present article is to decide the ideal number of warehouses for a food manufacturer in the north of Spain (Asturias) for the year 2000, and their ideal location in the Spanish Peninsula by cluster analysis. The stages followed are to comment first on the underlying assumptions of the study, then on the methodology and the structure of the program developed to solve the problem, as well as on their input and output files. How the cluster and cost are calculated is discussed and finally the solution to this real case is provided.

Systemic approach to supply chain management through the viable system model and the theory of constraints

Abstract In today’s environment, Supply Chain Management (SCM) takes a key role in business strategy. A major challenge is achieving high customer service level under a reasonable operating expense and investment. The traditional approach to SCM, based on local optimisation, is a proven cause of meaningful inefficiencies – e.g. the Bullwhip Effect – that obstruct the throughput. The systemic (holistic) approach, based on global optimisation, has been shown to perform significantly better. Nevertheless, it is not widely expanded, since the implementation of an efficient solution requires a suitable scheme. Under these circumstances, this paper proposes an integrative framework for supply chain collaboration aimed at increasing its efficiency. This is based on the combined application of the Beer’s Viable System Model (VSM) and the Goldratt’s Theory of Constraints (TOC). VSM defines the systemic structure of the supply chain and orchestrates the collaboration, while TOC implements the systemic behaviour – i.e. integrate processes – and define performance measures. To support this proposal, we detail its application to the widely used Beer Game scenario. In addition, we discuss its implementation in real supply chains, highlighting the key points that must be considered.