The broad aim of this research is to contribute knowledge that enables improvements in how designers model decision alternatives at the systems level—i.e., how they model different system configurations and concepts. There are three principal complications: (1) design concepts and systems configurations are partially-defined solutions to a problem that correspond to a large set of possible design implementations, (2) each concept or configuration may operate on different physical principles, and (3) decisions typically involve tradeoffs between multiple competing objectives that can include "non-engineering" considerations such as production costs and profits. This research is an investigation of a data-driven approach to modeling partially-defined system alternatives that addresses these issues. The approach is based on compositional strategy in which designers model a system alternative using abstract models of its components. The component models are representations of the rational tradeoffs available to designers when implementing the components. Using these models, designers can predict key properties of the final implementation of each system alternative. A new construct, called a parameterized efficient set, is introduced as the decision-theoretic basis for generating the component-level tradeoff models. Appropriate efficiency criteria are defined for the cases of deterministic and uncertain data. It is shown that the model composition procedure is mathematically sound under reasonable assumptions for the case of deterministic data. This research also introduces an approach for describing the valid domain of a data-driven model based on the use of support-vector machines. Engineering examples include performing requirements allocation for a hydraulic log splitter and architecture selection for a hybrid vehicle.

Thesis (Ph.D.)--Stanford University, 1992. / Includes bibliographical references (leaves 236-249).

Σκοπός της παρούσας διπλωματικής εργασίας είναι η ανάπτυξη ενός Συστήματος Υποστήριξης Αποφάσεων βασισμένο στα Ασαφή Γνωστικά Δίκτυα, το οποίο θα μπορεί να χρησιμοποιηθεί στον ιατρικό τομέα για τη διάγνωση ασθενειών και πιο συγκεκριμένα της νόσου του Parkinson. Αρχικά θα γίνει περιγραφή των Ασαφών Γνωστικών Δικτύων αλλά και του τρόπου υλοποίησης ενός Συστήματος Υποστήριξης Αποφάσεων για τη διάγνωση της νόσου του Parkinson. Στη συνέχεια θα γίνει πείραμα με στόχο να διαγνωσθεί το στάδιο, στο οποίο βρίσκονται τρεις ασθενείς. Αυτή η διάγνωση θα γίνει με δύο διαφορετικούς τρόπους. Τέλος θα συγκριθούν και θα αναλυθούν τα πειραματικά αποτελέσματα καθώς και τα συμπεράσματα που προκύπτουν από μία τέτοια έρευνα. / The purpose of this diploma thesis is to develop a Decision Support System based on Fuzzy Cognitive Maps. This system can be used in medicine in order to diagnose diseases, and more specifically Parkinson’s disease. After that three patients will be examined and the system will diagnose the stage of their disease. This diagnose will be achieved in two different ways. Finally we will compare and analyze the experimental results and the conclusions derived from such research.

Thesis (MSc)--Stellenbosch University, 2004. / ENGLISH ABSTRACT: Multi-criteria decision-making (MCDM) is a set of techniques designed specifically for the analysis of complex problems. Geographical Information Systems (GIS) focus on spatial problem-solving and spatial analysis. The integration of these methodologies offers a powerful approach to decision making. Despite the fact that most spatial decision problems are multi-criteria problems by nature, the process of MCDM is not well established or effectively integrated into the field of spatial analysis and GIS. This research focuses on bridging the gap between MCDM and GIS. To this end, a generic MCDM extension was designed and implemented in ArcView. As a result, a first version MCDM extension is offered. The extension expands ArcView's functionality with a limited set of MCDM methods. This functionality is illustrated on two problems involved with developing the tourism potential at Coutada 16 Wildlife Reserve in Mozambique. The MCDM extension facilitates procedures that allow the evaluation of spatial problems and includes the ability to deal with both raster and vector data. This system offers a generic problemsolving environment, which can be used to evaluate geographical problems of any nature. This research identifies a number of improvements to the developed functionality and successfully illustrates the potential problem-solving capabilities associated with MCDM integrated with ArcView. / AFRIKAANSE OPSOMMING: Multi Kriteria Besluitneming (MKBN) is n versameling metodes vir die analise van komplekse probleme. Geografiese Inligtingstelsels (GIS) fokus op geografiese probleemoplossing en analise. Die integrasie van hierdie twee metodologieë bied 'n kragtige benadering tot besluitneming. Ten spyte daarvan dat die meeste geografiese probleme in wese meerveranderlik van aard is, is MKBN nie effektiefbinne die raamwerk van GIS geïntegreer nie. Hierdie studie fokus op die oorbrugging van die gaping tussen MKBN en GIS. Met hierdie doel voor oë is 'n generiese MKBN-uitbreiding vir ArcView ontwerp en geïmplementeer. Die resultaat is 'n eerste- weergawe MKBN-uitbreiding. Die uitbreiding brei ArcView se funksionaliteit uit om 'n beperkte versameling MKBN-metodes in te sluit. Die nuut ontwikkelde funksies word geïllustreer aan die hand van twee probleme wat die ontwikkeling van die toerismepotensiaal vir die Coutada 16 Wildreservaat in Mosambiek aanspreek. Die uitbreiding maak voorsiening vir 'n MKBN-evaluasie van geografiese probleme en besit die vermoë om beide vektor- en roosterdata te analiseer. Hierdie stelsel verskaf 'n generiese omgewing vir probleemoplossing wat gebruik kan word om byna enige geografiese probleem te analiseer. Die studie identifiseer verbeteringe op en uitbreidings van die ontwikkelde funksies en slaag daarin om die potensiaal van probleemoplossing wat deur die integrasie van MKBN-tegnieke met ArcView moontlik gemaak word, te illustreer.

Thesis (MEng)--Stellenbosch University, 2014. / ENGLISH ABSTRACT: The demand for rail travel has increased over the years. As a result, it is becoming mandatory for railway industries to maintain very high availability of their assets to ensure that service levels are high. Railway industries require both their infrastructure and rolling stock assets maintained efficiently to sustain reliability. There has been on-going research on how maintenance can be carried out in a cost effective manner. However, the majority of this research has been done for infrastructure and the rolling stock maintenance has not been properly covered. The purpose of this research is to contribute to the maintenance sector of rolling stock for railway industries by developing a decision support model for rolling stock based on reliability and cost. The model is developed as an optimization problem of a system containing several components dependent on each other with different reliability characteristics. In this model, a mixed integer nonlinear problem is developed and solved using an exact method and metaheuristics methods. The Metrorail facility in Cape Town was chosen as a case study. Failure history and cost data were gathered from the facility and the information was applied to the model developed. The case study was investigated and different results were achieved using both exact and metaheuristics methods. The final result from the study is an optimal maintenance schedule based on reliability and cost. The developed model serves as a practical tool railway companies can adopt to schedule rolling stock maintenance to achieve a high level of reliability and at the same time maintaining minimum cost expenditure. / AFRIKAANSE OPSOMMING: Die vraag na spoorvervoer het oor die jare toegeneem. Dus het dit belangrik geword dat die spoorweg se bates hoogs toeganklik moet wees om te verseker dat die vlak van dienslewering hoog bly. Die spoorweg industrie besef dat hulle infrastruktuur, lokomotiewe, waens ens. effektief in stand gehou moet word sodat dit betroubaar kan wees. Navorsing word nog steeds gedoen oor hoe instandhouding op ’n koste-effektiewe wyse gedoen kan word. Die meeste van hierdie navorsing gaan egter oor infrastruktuur en instandhouding word nie ordentlik gedek nie. Die doel met hierdie navorsing is om by te dra tot die instandhoudingsektor van die spoorweg deur om ’n besluit-ondersteunende model vir lokomotiewe, waens, ens wat op betroubaarheid en koste gegrond is, te ontwikkel. Die model is ontwikkel as ’n optimasie probleem van ’n sisteem wat verskillende komponente wat van mekaar afhanklik is maar oor verskillende betroubaarheidskenmerke beskik, inluit. In hierdie model word ’n gemengde, heeltal nie-lineêre probleem ontwikkel en met ’n eksakte metode en metaheuristiese metodes opgelos. Die Metrorail fasiliteit in Kaapstad is vir die gevalle studie gekies. Die geskiedenis van mislukkings en koste data is by die fasiliteit versamel en die inligting is op die model wat ontwikkel is, toegepas. Die gevalle studie is ondersoek, en verskillende resultate is met eksakte en metaheuristiese metodes bereik. Die finale uitkomste van die studie is ’n optimale instandhoudingskedule wat op betroubaarheid en koste gegrond is. Die model wat ontwikkel is dien as ’n praktiese instrument wat spoormaatskappye kan gebruik om die instandhouding van lokomotiewe, waens ens. te reël sodat ’n hoë vlak van betroubaarheid bereik kan word en kostes terselfdertyd tot ’n minimum beperk kan word.

Thesis (MEng)--Stellenbosch University, 2014. / ENGLISH ABSTRACT: Effective management of spare parts inventory is essential to companies because it influences inventory costs and asset utilization. The vast and diverse portfolio of spare parts, intermittent demand patterns and contradicting objectives between departments are examples of some of the factors that complicate Spare Parts Management (SPM). Managers of spare parts are faced with trade-off decisions between risk and cost on a daily basis. These decisions include, amongst many, determining appropriate stock levels and order frequencies. Despite the importance of SPM, decisions are however often made intuitively in practice with little factual support, and the decision-making process is commonly constrained within departmental silos. Even though there is a large body of academic knowledge on this topic, practical applications of spare parts inventory solutions lag behind theoretical studies. The majority of studies in literature focus on single components of SPM, such as demand forecasting and parts classification, whereas fewer studies consider the decision-making process itself. This study proposes a decision-making framework for spare parts inventory management. The framework is based on a wide-ranging literature review that focuses on capturing the essence of Spare Parts Management (SPM), but also acknowledges the interconnectedness of the problem. Therefore, core inventory management principles, as well as closely related topics such as Supply Chain Management (SCM) and Physical Asset Management (PAM), are studied in the context of spare parts. The broad scope of the literature study leads to a holistic approach to the problem and prevents sub-optimization. The proposed framework condenses principles from various fields of study (SCM, PAM, Classification and Inventory Management) into a stepwise methodology presented as a decision-making framework. The objective of the framework is to provide managers with a structured process, based on factual information, to enable better decision-making in the field. Furthermore, the framework aims to capture the fundamentals of SPM in a simplistic manner to ease the adoption of the framework in practice. A case study is conducted in the South African mining industry to validate the framework. The case study demonstrates that the framework is practical, provides structured guidance, and assists managers to make trade-off decisions in managing spare parts inventory. / AFRIKAANSE OPSOMMING: Effektiewe voorraadbestuur van onderdele is belangrik vir maatskappye omdat dit voorraadkoste en die benutting van bates beïnvloed. Die bestuur van onderdele is ’n komplekse probleem. Ondermeer is die portefeulje van onderdele items breed en divers, die vraagpatrone sporadies en word die voorraadvlakke geaffekteer deur kontrasterende doelwitte tussen verskillende departemente. Bestuurders van onderdele word daagliks gekonfronteer met besluite rakende risiko’s en kostes, soos om toepaslike voorraadvlakke te bepaal en om te besluit wanneer om bestellings te plaas. Hierdie besluite word dikwels intuïtief geneem met min feitelike ondersteuning en insette in die besluitnemingsproses word gereeld beperk tot sekere departemente. Ten spyte van die geweldige akademiese belang in die onderwerp, is daar min suksesvolle praktiese toepassings. Die meerderheid van studies in die literatuur fokus op spesifieke elemente van onderdele bestuur, soos vooruitskatting en klassifisering van parte, terwyl minder op die besluitnemingsproses konsentreer. Hierdie studie stel ’n besluitnemingsraamwerk vir die bestuur van onderdele voorraad voor. Die raamwerk is gegrond op ’n deeglike literatuurstudie wat die essensie van onderdele bestuur ondersoek, maar ook die interverbondenheid van die probleem in ag neem. Voorraadbestuurbeginsels en verwante onderwerpe soos Voorsieningskettingbestuur en Fisiese Batebestuur word dus bespreek. Die breë omvang van die literatuurstudie lei tot ’n holistiese benadering wat sub-optimering van die probleem voorkom. Die voorgestelde raamwerk som beginsels uit verskillende relevante studievelde op in ’n stapsgewyse metode wat voorgestel word as ’n besluitnemingsraamwerk. Die doel van die raamwerk is om bestuurders te voorsien met ’n gestruktureerde proses, gebaseer op feitelike inligting, om besluitneming in die veld te verbeter. Verder poog die raamwerk om die fundamentele konsepte in voorraadbestuur vas te vang in ’n eenvoudige manier sodat die raamwerk maklik geïmplementeer kan word in die praktyk. Die voorgestelde raamwerk is gevalideer deur middel van ’n gevallestudie in die Suid-Afrikaanse mynbedryf. Die gevallestudie toon dat die voorgestelde raamwerk prakties is, die besluitnemingsproses op ’n gestruktureerde wyse lei, en bestuurders help om beter, ingeligte besluite te neem.

Este estudo teve como objetivo propor uma metodologia para a classificação de alternativas de abertura da inovação em Pequenas e Médias Empresas (PMEs). A estruturação da metodologia passou por duas etapas, a primeira relacionada à base teórica e a segunda à base matemática. Na base teórica foi elaborado um portfólio bibliográfico sobre o tema que forneceu base às análises bibliométricas e sistemática. Os resultados da análise bibliométrica revelaram que a literatura sobre o tema Inovação Aberta (IA) em PMEs, apesar de crescente, ainda não está consolidada. Por meio da análise sistemática do conteúdo dos artigos foram identificadas 18 alternativas e 29 variáveis, que foram agrupadas em 4 critérios, relacionados ao processo de abertura de inovação nas PMEs. Na base matemática, segunda etapa, foi estruturada a metodologia por meio das etapas previstas pelo método ELECTRE TRI. A metodologia conta com oito etapas que possibilitam uma classificação das alternativas de forma estruturada, de acordo com as classes pré-definidas. Posteriormente, a metodologia foi implementada, para teste, em 3 empresas de diferentes setores da economia. Os resultados apontaram diferentes níveis de desempenho nos critérios relacionados às capacidades de cada empresa, sendo: 2,57; 2,37 e; 1,74 para as empresas 1; 2 e; 3, respectivamente, em uma escala de 0 a 4. O resultado da classificação das alternativas de abertura da inovação alocou na classe “A”, considerada a classe mais favorável às empresas, 50% das alternativas para a Empresa 1, 28% para a Empresa 2 e 11% para a Empresa 3. Ao se considerar um cenário de melhoria, foi implementado um aumento de 10% no peso do critério crítico de cada empresa. Com esta variação foram alocadas 72% das alternativas na classe “A” para a Empresa 1, e 22% para a Empresa 3. Para a Empresa 2 não houve reclassificação. Com estes resultados, pôde-se verificar que o aumento no nível de capacidade das empresas pode levar a um aumento significativo de alternativas realocadas em classes superiores. / This study aimed to propose a methodology for the classification of alternatives for opening innovation in Small and Medium Enterprises (SMEs). The structuring of the methodology went through two phases, a first relation to the theoretical basis and a second mathematical basis. On the theoretical basis a bibliographic portfolio was elaborated on the subject that serves as the basis for bibliometric and systematic analyzes. The results of the bibliographic analysis reveal that the literature on topic AI in SMEs, although increasing, is not yet consolidated. Through the systematic analysis of the content of the identified articles 18 alternatives and 29 variables were grouped into 4 criteria related to the process of opening innovation in SMEs. In the mathematical base, second stage, a methodology was structured through the ELECTRE TRI method steps. The methodology has the steps that allow a classification of the alternatives in a structured way, according to the predefined classes. Subsequently, the methodology was implemented, for testing, in 3 companies from different sectors of the economy. The results indicate the different levels of performance in the following criteria: 2.57; 2.37 e; 1.74 for enterprises 1; 2 e; 3, respectively, on a scale of 0 to 4. The result of the classification of alternatives to open innovation in class "A", considered a class more favorable to the company, 50% of the alternatives for the company 1, 28% for the Company 2 and 11% for Company 3. When considering a scenario of improvement, a 10% increase in the weight of the companies' critical criteria was implemented. With this variation 72% of the alternatives for class "A" were adopted for company 1, and 22% for company 3. With this, it can be verified that the increase there is no level of capacity of the companies can lead to an increase Significant reallocated alternatives in upper classes.

O incentivo ao uso racional de medicamentos fator importante na melhoria das condições de saúde. A efetivação desta racionalidade, no entanto, esbarra tanto em fatores culturais como em fatores de escassez ou mesmo ausência de informação confiável. Os farmacêuticos, membros indispensáveis de qualquer equipe multidisciplinar de saúde, cumprem papel preponderante na disseminação deste conhecimento especializado. Com o objetivo de suprir as lacunas existentes quanto a qualidade da informação farmacológico-terapêutica existente no Brasil, este trabalho desenvolve um sistema de informações sobre medicamentos cardiovasculares aliado a um Sistema de Suporte a Decisão Baseado em Alertas, utilizando uma arquitetura que combina bases de dados relacionais com Medical Logical Modules. O sistema desenvolvido proporciona, a farmacêuticos e outros profissionais da área da saúde, não apenas consultas a informação, mas também sugestões e alertas contextuais referentes ao uso correto de medicamentos, contribuindo, assim, para o treinamento no próprio trabalho destes. / The rational use of drugs is an important factor to the health conditions improvement. The pharmacists are responsible for advising customers about the rational use of drugs, but this professional effort is limited by cultural factors, absence or lack of reliable infomation. The aim of this work is to support pharmacists through a cardiovascular medicine Information System attached to an Alert-based Decision Support System whose structure combines Medical Logical Modules and relational databases. The developed system not only provides the health professional and pharmacist with drug information, but also suggests or alerts them about the proper use of drugs as well it contributes to their in site training.

A regulação de consultas especializadas tem se mostrado como uma das áreas mais problemáticas do Sistema Único de Saúde (SUS) no Brasil. Cabe aos gestores de saúde nos municípios, estados, e governo federal, estabelecerem mecanismos de regulação coerentes com o volume de recursos disponíveis e com o contingente populacional a atender. Diversas centrais de regulação para atendimentos especializados foram implantadas nas secretarias municipais de saúde e sistemas de informação foram criados como ferramentas para apoio a estas centrais. Seu escopo tem sido progressivamente ampliado, de maneira a incluir uma visão crítica das necessidades da população em relação à capacidade de atendimento dos prestadores de serviço. No processo de regulação de consultas especializadas, duas questões têm-se destacado: (1) para um dado caso, quais pacientes têm maior prioridade de atendimento, e (2) quais prestadores de serviço podem resolver melhor o caso? Fundamentado nestas duas questões, e a partir da consideração dos requisitos legitimados na área da assistência à saúde, este trabalho propõe um sistema para apoio à decisão de agendamento de consultas especializadas para servir às centrais de regulação. O sistema proposto integra análise de decisão multi-critério e programação linear para o agendamento das consultas, onde a alocação dos pacientes é definida em função da relevância relativa de um conjunto de critérios relacionados à noção de efetividade da assistência médica especializada e da capacidade de atendimento das unidades de assistência credenciadas. Da integração destes modelos resulta uma representação que leva em conta simultaneamente os aspectos relacionados ao diagnóstico médico e suas conseqüências na vida do paciente, os aspectos relacionados às instalações e processos disponíveis nas unidades assistenciais credenciadas, e os aspectos relacionados à dificuldade de acesso do paciente a estas unidades. O uso do sistema permite que as informações pessoais e médicas do paciente, assim como as informações sobre as unidades assistenciais, sejam incorporadas em um modelo de programação linear de maneira a maximizar a efetividade do conjunto de solicitações para cada especialidade. Os modelos foram implementados em um sistema informatizado, e aplicados em uma parcela dos serviços da Secretaria Municipal de Saúde de Porto Alegre para as especialidades de cardiologia e cirurgia vascular. O sistema e os resultados obtidos foram validados por um grupo de peritos, que confirmou a viabilidade do uso deste modelo como uma ferramenta para a otimização da alocação de recursos no atendimento especializado pelo SUS. / The regulation of specialized medical consultations has been one of the most problematic areas of the Government Unified Health System (SUS) in Brazil. It is the role of health managers from cities, states and federal government to establish coherent mechanisms of regulation with the amount of available resources and the population contingent to be assisted. Many regulation centers for specialized consultations had been created in public city health departments, and many information systems were developed to support these centers. Its target has been gradually extended, in a way to include a critical vision of the necessities of the population in relation to the capacity of attendance of the service rendering. On the specialized consultations regulatory process, two questions arise: (1) for a random situation, which patient has priority to be assisted? (2) Which health providers can better solve this problem? Based on these two questions, and from the consideration of the legitimated requirements in health care, this work considers a decision support system for the scheduling process of specialized consultations into regulation central offices. The considered system integrates multi-criteria analysis and linear programming for the scheduling process, where the allocation of the patients is defined in function of the relative relevance of a set of criteria related to the notion of effectiveness of the specialized medical assistance and the capacity of assistance of the credential service providers. By the integration of these models, a representation results that simultaneously takes in account the related aspects to the medical diagnosis and its consequences in the patient’s life, the aspects related to the installations and available processes in the credential assistance units, and the aspects related to the difficulty of access of the patient to these units. The use of the system allows that the personal and medical information of the patient, as well as the information on the assistance units, are incorporated in a model of linear programming in a way to maximize the effectiveness of the set of requests for each specialty. The models had been implemented in a decision support system, and applied in a parcel of the services of the Porto Alegre City Health Department for cardiology and vascular surgery. The information system and the outgoing results were validated by a group of experts, which confirmed the model viability for using as a tool to improve the resource distribution at the SUS's specialized assistance.

Clinical diagnosis of chronic disease is a vital and challenging research problem which requires intensive clinical practice guidelines in order to ensure consistent and efficient patient care. Conventional medical diagnosis systems inculcate certain limitations, like complex diagnosis processes, lack of expertise, lack of well described procedures for conducting diagnoses, low computing skills, and so on. Automated clinical decision support system (CDSS) can help physicians and radiologists to overcome these challenges by combining the competency of radiologists and physicians with the capabilities of computers. CDSS depend on many techniques from the fields of image acquisition, image processing, pattern recognition, machine learning as well as optimization for medical data analysis to produce efficient diagnoses. In this dissertation, we discuss the current challenges in designing an efficient CDSS as well as a number of the latest techniques (while identifying best practices for each stage of the framework) to meet these challenges by finding informative patterns in the medical dataset, analysing them and building a descriptive model of the object of interest and thus aiding in medical diagnosis. To meet these challenges, we propose an extension of conventional clinical decision support system framework, by incorporating artificial immune network (AIN) based hyper-parameter optimization as integral part of it. We applied the conventional as well as optimized CDSS on four case studies (most of them comprise medical images) for efficient medical diagnosis and compared the results. The first key contribution is the novel application of a local energy-based shape histogram (LESH) as the feature set for the recognition of abnormalities in mammograms. We investigated the implication of this technique for the mammogram datasets of the Mammographic Image Analysis Society and INbreast. In the evaluation, regions of interest were extracted from the mammograms, their LESH features were calculated, and they were fed to support vector machine (SVM) and echo state network (ESN) classifiers. In addition, the impact of selecting a subset of LESH features based on the classification performance was also observed and benchmarked against a state-of-the-art wavelet based feature extraction method. The second key contribution is to apply the LESH technique to detect lung cancer. The JSRT Digital Image Database of chest radiographs was selected for research experimentation. Prior to LESH feature extraction, we enhanced the radiograph images using a contrast limited adaptive histogram equalization (CLAHE) approach. Selected state-of-the-art cognitive machine learning classifiers, namely the extreme learning machine (ELM), SVM and ESN, were then applied using the LESH extracted features to enable the efficient diagnosis of a correct medical state (the existence of benign or malignant cancer) in the x-ray images. Comparative simulation results, evaluated using the classification accuracy performance measure, were further benchmarked against state-of-the-art wavelet based features, and authenticated the distinct capability of our proposed framework for enhancing the diagnosis outcome. As the third contribution, this thesis presents a novel technique for detecting breast cancer in volumetric medical images based on a three-dimensional (3D) LESH model. It is a hybrid approach, and combines the 3D LESH feature extraction technique with machine learning classifiers to detect breast cancer from MRI images. The proposed system applies CLAHE to the MRI images before extracting the 3D LESH features. Furthermore, a selected subset of features is fed to a machine learning classifier, namely the SVM, ELM or ESN, to detect abnormalities and to distinguish between different stages of abnormality. The results indicate the high performance of the proposed system. When compared with the wavelet-based feature extraction technique, statistical analysis testifies to the significance of our proposed algorithm. The fourth contribution is a novel application of the (AIN) for optimizing machine learning classification algorithms as part of CDSS. We employed our proposed technique in conjunction with selected machine learning classifiers, namely the ELM, SVM and ESN, and validated it using the benchmark medical datasets of PIMA India diabetes and BUPA liver disorders, two-dimensional (2D) medical images, namely MIAS and INbreast and JSRT chest radiographs, as well as on the three-dimensional TCGA-BRCA breast MRI dataset. The results were investigated using the classification accuracy measure and the learning time. We also compared our methodology with the benchmarked multi-objective genetic algorithm (ES)-based optimization technique. The results authenticate the potential of the AIN optimised CDSS.