Soft Computing for Sustainability Science

Sustainability Science is a research field that seeks to understand the fundamental character of interactions between Nature and Society. Because of the high degree of complexity of the problems and challenges it faces, this field requires new methodological approaches, tools and techniques that enable decision-makers and stakeholders can evaluate and make decisions based on a wide range of uncertainty and little information. It is here where Soft Computing methodologies can play an important role in addressing many of these challenges. The inherent tolerance of uncertainty and imprecision, and the robustness of the techniques that make up this paradigm can help solve or reduce the impact of these problems. This chapter introduces this book as a a catalogue of the most successful Soft Computing methodologies applied to Sustainability Science.

The shipping industry is today increasingly concerned with challenges related with sustainability. CO\(_2\) emissions from shipping, although they today contribute to less than 3% of the total anthropogenic emissions, are expected to rise in the future as a consequence of increased cargo volumes. On the other hand, for the 2 \(^\circ \)C climate goal to be achieved, emissions from shipping will be required to be reduced by as much as 80% by 2050. The power required to propel the ship through the water depends, among other parameters, on the trim of the vessel, i.e. on the difference between the ship’s draft in the fore and the aft of the ship. The optimisation of the trim can, therefore, lead to a reduction of the ship’s fuel consumption. Today, however, the trim is generally set to a fixed value depending on whether the ship is sailed in loaded or ballast conditions, based on results performed on model tests in basins. Nevertheless, the on-board monitoring systems, which produce a huge amount of historical data about the life of the vessels, lead to the application of state of the art data analytics techniques. The latter can be used to reduce the vessel consumption by means of optimising the vessel operational conditions. In this book chapter, we present the potential of data-driven based techniques for accurately predicting the influence of independent variables measured from the on board monitoring system and the fuel consumption of a specific case study vessel. In particular, we show that gray-box models (GBM) are able to combine the high prediction accuracy of black-box models (BBM) while reducing the amount of data required for training the model by adding a white-box model (WBM) component. The resulting GBM model is then used for optimising the trim of the vessel, suggesting that between 0.5 and 2.3% fuel savings can be obtained by appropriately trimming the ship, depending on the extent of the range for varying the trim.

This chapter presents a spatial decision support system called FuzzyCovering, which is designed to support the decision-making process related to facility location problems. Different components that facilitate modeling, the solution and results display, specifically about covering location problems are integrated in FuzzyCovering. FuzzyCovering allows the study of various scenarios of facilities location and provides a range of solutions that allow the users to make the best decisions. To treat the uncertainty inherent to some underlying parameters of the real location problems, FuzzyCovering integrates a fuzzy approach in which the problem constraints can be imprecisely defined. A detailed description of the architecture and functionality of the system is presented, and a simulated practical case of a maximal covering location problem with fuzzy constraints is shown to demonstrate the benefits of FuzzyCovering.

Locating theory seeks to exploit geographic information in order to identify the best suited areas to place new facilities. This chapter tackles an optimization problem aimed at finding suitable locations for a new infrastructure in Spain by considering sustainability criteria and from a fuzzy perspective. In order to solve this problem, open georeferenced data provided through a Geographic Information System are used. Several fuzzy membership functions are proposed to represent the level of possible membership to the desired locations defined by the problem criteria. The resulting locations are appropriately combined under different choices of a given decision maker. In addition, the approach can be easily adapted to tackle similar location problems.

Spectroscopic techniques have shown great potential due to their quick response, cost-effective, non-destructive and non-invasive nature, and environmental friendliness. These characteristics make this technology very attractive for sustainable industry and research activities, being viticulture industry no exception. Spectroscopic techniques are an appealing alternative for ripeness assessment and harvest date determination as well as for plant variety and clone determination. Numerous recent works have clearly demonstrated that it is highly advantageous to process the high dimensionality spectroscopic data with soft computing or multivariate analysis techniques such as Partial Least Squares, Neural Networks or Support Vector Machines. In this review, focus will be given to two emergent subfields in viticulture where the combination of spectroscopy and soft computing is fundamental: (1) The difficult measurement of enological parameters, namely sugar content, pH and anthocyanin content, in samples containing a small number of grape berries, with the aim of assessing grapes’ ripeness; (2) The multiclass problem of identifying plant varieties and clones. The results of the various works in these subfields will be presented. The present article starts with a brief description of the spectroscopy principles and continues by making an overview of the scientific literature considering the number of berries per sample and the total number of samples in the various works. The use of different varieties, vintages and harvest locations in the same model will also be addressed. Special attention is given to the validation methods employed and to algorithm comparison. Some suggestions are presented in order to facilitate future comparison of published results.

With the rollout of smart metering infrastructure at large scale, demand-response programs may now be tailored based on consumption and production patterns mined from sensed data. In previous works, groups of similar energy consumption profiles were obtained. But, discovering typical consumption profiles is not enough, it is also important to reveal various preferences, behaviors and characteristics of individual consumers. However, the current approaches cannot determine clusters of similar consumer or prosumer households. To tackle this issue, we propose to model the consumer clustering problem as a multi-instance clustering problem and we apply a multi-instance clustering algorithm to solve it. We model a consumer as a bag and each bag consists of instances, where each instance will represent a day or a month of consumption. Internal indices were used for evaluating our clustering process. The obtained results are general applicable, and will be useful in a general business analytics context.

The problem of ranking a set of technology packages that are best suited for growing crops, is developed with a multicriteria group decision model. The group decision model is based on ELECTRE GD, a group decision method for multicriteria ranking problems, strongly based on ELECTRE III, developed to work on those cases where there is great divergence among the decision-makers. We use a practical case study to show our approach, where a group of decision-makers evaluates among the available technology packages to an agricultural company, in order to select the most appropriate alternative. The proposed model generates an agreed collective solution that aids those decision-makers with different interests, to reach (through an iterative process) an agreement on how to rank the technology packages. The proposed procedure is also based on a preference disaggregation approach for reaching agreement between individuals. To support the proposal of a temporary collective solution, individual inter-criteria parameters are inferred concerning individual and global preference for outranking methods in a feedback process.

Impact investing is an investment practice that is characterized by the explicit intentionality of attaining a social impact and the requisite of report and measure this impact in a transparent way. The investment decision making process has two main stages. In the first stage, filters are applied regarding four critical issues: target geography, impact theme, asset class and target return category. In this phase, the set of possible investment alternatives are determined based on their appropriateness for impact investment in terms of those four essential aspects. In a second stage, efficient portfolios are obtained taking into account financial criteria (maximizing expected return, minimizing risk) and trying to maximize the social impact of the portfolio of investments. In this chapter, we will focus on the establishment of the target geography for the impact investment proposing a fuzzy indicator of the appropriateness of a geographic area in terms of impact investment. This indicator will be based on Soft Computing techniques which are an attractive tool given the imprecise, ambiguous and uncertain nature of data related to social impact investment.

Vehicular Ad-Hoc Networks (VANETs) have attracted a high interest in recent years due to the huge number of innovative applications that they can enable. Some of these applications can have a high impact on reducing Greenhouse Gas emissions produced by vehicles, especially those related to traffic management and driver assistance. Many of these services require disseminating information from a central server to a set of vehicles located in a particular region. This task presents important challenges in VANETs, especially when it is made at big scale. In this work, we present a new approach for information dissemination in VANETs where the structure of the communications is configured using a model based on Covering Location Problems that it is optimized by means of a Genetic Algorithm. The results obtained over a realistic scenario show that the new approach can provide good solutions for very demanding response times and that obtains competitive results with respect to reference algorithms proposed in literature.

The growth of the Internet has fuelled the availability of e-commerce marketplaces and search engines must face with a huge amount of ambiguity and inconsistencies in the data. Product matching aims at disambiguating descriptions of products belonging to different websites in order to be able to recognize identical products and to merge the content from those identical items. In this work first we evaluate some similarity measures for string matching and then, we apply a complete product matching methodology to the retail market of used cars. We use a reference or master list of items and information about a wide variety of used cars offers. The resulting linkage allows energy efficiency assignment of the model identified.

In this chapter the design of a fault diagnosis system using fuzzy clustering techniques for a BKZ-340-140 29M steam generator in a thermoelectric power station is presented. The application aims to study the advantages of these techniques in the development of a fault diagnosis method with the characteristic to be robust to external disturbances and sensitive to small magnitude faults. The wavelet transform (WT) is used for isolating noise present in measurements. The fault diagnosis system was designed for the water-steam circuit of the steam generator by its great incidence in the correct operation of the generation blocks. The obtained results indicate the feasibility of the proposal.

Watershed identification is one of the main areas of study in the field of topography. It is critical in countless applications including sustainability and flood risk evaluation. Beyond its original conception, the watershed algorithm has proved to be a very useful and powerful tool in many different applications beside topography, such as image segmentation. Although there are a few publications reviewing the state-of-the-art of watershed algorithms, they are now outdated. In this chapter we review the most important works done on watershed algorithms, including the problem over-segmentation and parallel approaches. Open problems and future work are also investigated.

This chapter presents studies about the data imputation to estimate missing values, and the Data Editing and Imputation process to identify and correct values erroneously. Artificial Neural Networks and Support Vector Machines are trained as Machine Learning techniques on real and simulated data sets obtaining a complete data set what help to improve the quality of the variables that define the official indicators of the eight Millennium Development Goals.

This work shows a review about the multimodal transport network problem. This kind of problem has been studied for several researchers who look for solutions to the large numbers of problems relating on the transport systems like: traffic jam, pollution, delays, among others. In this work are presented a standard mathematical formulation for this problem and some other variations, which make the problem more complex and harder to be solved. There are many approaches to solve it that are found in the literature and they are divided according to classical methods and soft computing methodologies, which combine approximate reasoning as fuzzy logic and functional as metaheuristics and neural networks. Each approach has its advantages and disadvantages that are also shown. A novel approach to solve the multimodal transport network problem in fuzzy environment is developed and this approach is also applied in a theoretical problem to illustrate its effectiveness.

Maritime port operations usually comprise a spread spectrum of environmental challenges, which are often unique to each port site. For this reason, the Environmental Impact Assessment (EIA) process has been developed for evaluating the impact of port operations on the environment, including its natural, social and economic aspects. In this chapter, an EIA model based on a 2-tuple linguistic model is proposed to assess the overall environmental impact of Moa Port in Cuba by using a double matrix that represents impacts, which are characterized by multiple-criteria. The environmental impacts of factors and actions are also ranked from the most to the least risky. This EIA 2-tuple linguistic based model facilitates the handling of inherent uncertainty of criteria involved in an EIA problem by simplifying the sophisticated structure of the problem under consideration while computations are made without loss of information and provide a high interpretability of the EIA results.