José Parreño

Forecasting next-day price of electricity in the Spanish energy market using artificial neural networks

In this paper, next-day hourly forecasts are calculated for the energy price in the electricity production market of Spain. The methodology used to achieve these forecasts is based on artificial neural networks, which have been used successfully in recent years in many forecasting applications. The days to be forecast include working days as well as weekends and holidays, due to the fact that energy price has different behaviours depending on the kind of day to be forecast. Besides, energy price time series are usually composed of too many data, which could be a problem if we are looking for a short period of time to reach an adequate forecast. In this paper, a training method for artificial neural nets is proposed, which is based on making a previous selection for the multilayer perceptron (MLP) training samples, using an ART-type neural network. The MLP is then trained and finally used to calculate forecasts. These forecasts are compared to those obtained from the well-known Box-Jenkins ARIMA forecasting method. Results show that neural nets perform better than ARIMA models, especially for weekends and holidays. Both methodologies calculate more accurate forecasts-in terms of mean absolute percentage error-for working days that for weekends and holidays. Agents involved in the electricity production market, who may need fast forecasts for the price of electricity, would benefit from the results of this study.

LEARNING-BASED SCHEDULING OF FLEXIBLE MANUFACTURING SYSTEMS USING SUPPORT VECTOR MACHINES

Dispatching rules are usually applied to dynamically schedule jobs in flexible manufacturing systems (FMSs). Despite their frequent use a significant drawback is that the performance level of the rule is dictated by the current state of the manufacturing system. Because no rule is better than any other for every system state, it would be highly desirable to know which rule is the most appropriate for each given condition. To achieve this goal we propose a scheduling approach using support vector machines (SVMs). By using this technique and by analyzing the earlier performance of the system, “scheduling knowledge” is obtained whereby the right dispatching rule at each particular moment can be determined. Simulation results show that the proposed approach leads to significant performance improvements over existing dispatching rules. In the same way it is also confirmed that SVMs perform better than other traditional machine learning algorithms as the inductive learning when applied to FMS scheduling problem, due to their better generalization capability.

Applying a fuzzy logic methodology to waiting list management at a hospital emergency unit: a case study

This paper studies care provision at hospital emergency departments. It begins with a traditional approach, based on queuing theory and simulation models, and then applies a fuzzy approach to the system so as to analyse the certainty levels provided by the key parameters of the system. A real (practical) case that occurred at an emergency department is analysed using both methods. Finally, the comparative results between the traditional model and the proposed one are shown.

Intelligent decision support system for real-time water demand management

AbstractEnvironmental and demographic pressures have led to the current importance of Water Demand Management (WDM), where the concepts of efficiency and sustainability now play a key role. Water must be conveyed to where it is needed, in the right quantity, at the required pressure, and at the right time using the fewest resources. This paper shows how modern Artificial Intelligence (AI) techniques can be applied on this issue from a holistic perspective. More specifically, the multi-agent methodology has been used in order to design an Intelligent Decision Support System (IDSS) for real-time WDM. It determines the optimal pumping quantity from the storage reservoirs to the points-of-consumption in an hourly basis. This application integrates advanced forecasting techniques, such as Artificial Neural Networks (ANNs), and other components within the overall aim of minimizing WDM costs. In the tests we have performed, the system achieves a large reduction in these costs. Moreover, the multi-agent environme...

Application of Genetic Algorithms to Container Loading Optimization

—Standardization of transport means, such as, containers has a direct impact on the transportation efficiency sought by European transport policies. In this paper, we present a genetic algorithm application to the container loading problem trying to maximize the cargo volume accommodated in the container whilst ensuring that loading restrictions are met, and thus achieving a reduction in the number of freight to hire and thereby a reduction in costs. The proposed method has been compared to similar models, and the results obtained are similar or even improved.

A case-study about shift work management at a hospital emergency department with genetic algorithms

Organising shifts, or work rosters, is a problem that affects a large number of businesses where staff are subject to some kind of work rotation or other. Researchers in the fields of Operations Research and Artificial Intelligence have devised several systems in an attempt to optimise the problem. This paper focuses on the problem of medial staff shift rotation at a hospital emergency department, and attempts to establish a method to automate its management by applying genetic algorithms. It also analyses one of the duty rosters that came out of the proposed solution.

Rural tourism in Spain: natural resources as sources of competitive advantage

The aim of this paper is to analyse the state of the art of the Spanish rural tourism sector, as well as performing forecasts for this strategically important sector of Spanish economy. Section 1 of the paper describes rural tourism in Spain, while in Section 2 three time series belonging to this sector are analysed, and then forecasts are calculated by applying Box-Jenkins and Artificial Neural Nets methodologies. Finally, the paper summarises major conclusions and implications for policy makers and managers involved in rural tourism in Spain.

Agents Playing the Beer Distribution Game: Solving the Dilemma Through the Drum-Buffer-Rope Methodology

The Beer Distribution Game (BDG) is a widely used experiential learning simulation game aimed at teaching the basic concepts around Supply Chain Management (SCM). The goal in this problem is to minimize inventory costs while avoiding stock-outs –hence the players face the dilemma between storage and shortage. Human players usually get confused giving rise to significant inefficiencies in the supply chain, such as the Bullwhip Effect. This research paper shows how artificial agents are capable of playing the BDG effectively. In order to solve the dilemma, we have integrated supply chain processes (i.e. a collaborative functioning) through the Drum-Buffer-Rope (DBR) methodology. This technique, from Goldratt’s Theory of Constraints (TOC), is based on bottleneck management. In comparison to traditional alternatives, results bring evidence of the great advantages induced in the BDG by the systems thinking. Both the agent-based approach and the BDG exercise have proved to be very effective in illustrating managers the underlying structure of supply chain phenomenon.

Reverse Logistics: A business opportunity in time of crisis

The current project is aimed at analyzing the possibilities of the Reverse Logistics as a new market niche in industry through the creation of a standard industrial estate that clusters several activities related to the industrial waste appreciation and that can be benefited from the economies of scale, from the synergies obtained between the companies settled there and the collaboration of public authorities. The results obtained show that the initiative also represents a business opportunity for the promoters, when there is a serious crisis in the sector, especially in Spain.