Design Science Research Framework for Performance Analysis Using Machine Learning Techniques

Enhancing medical radiation science education through a design science research methodology

Course and instructor evaluations (CIE) are essential tools for assessing educational quality in higher education. However, traditional CIE systems often suffer from inconsistencies between structured responses and open-ended feedback, leading to unreliable insights and increased administrative workload. This study suggests a modular system to address these challenges, leveraging sentiment analysis and inconsistency detection to enhance the reliability and efficiency of CIE processes.BackgroundImproving the reliability of CIE data is crucial for informed decision-making in higher education. Existing methods fail to address discrepancies between numerical scores and textual feedback, resulting in misleading evaluations. This study proposes a system to identify and exclude inconsistent data, providing more reliable insights.MethodsUsing the Design Science Research Methodology (DSRM), a system architecture was developed with five modules: data collection, preprocessing, sentiment analysis, inconsistency detection, and reporting. A dataset of 13,651 anonymized Turkish CIE records was used to train and evaluate machine learning algorithms, including support vector machines, naive Bayes, random forest, decision trees, K-nearest neighbors, and OpenAI’s GPT-4 Turbo Preview model. Sentiment analysis results from open-ended responses were compared with structured responses to identify inconsistencies.ResultsThe GPT-4 Turbo Preview model outperformed traditional algorithms, achieving 85% accuracy, 88% precision, and 95% recall. Analysis of a prototype system applied to 431 CIEs identified a 37% inconsistency rate. By excluding inconsistent data, the system generated reliable reports with actionable insights for course and instructor performance. The purpose of this study is to design and evaluate a new system using the Design Science Research (DSR) approach to enhance the accuracy and reliability of course evaluation processes employed in higher education institutions. The modular system effectively addresses inconsistencies in CIE processes, offering a scalable and adaptable solution for higher education institutions. By integrating advanced machine learning techniques, the system enhances the accuracy and reliability of evaluation reports, supporting data-driven decision-making. Future work will focus on refining sentiment analysis for neutral comments and broadening the system’s applicability to diverse educational contexts. This innovative approach represents a significant advancement in leveraging technology to improve educational quality.

Natural science research is the most common and widely used research methodology. Various fields such as physics, biology, social and behavioral apply this research methodology. However, not all research is suitable for using the natural science research methodology in computer fields such as information technology, computer science, and information systems. Research that aims to solve problems by developing an artefact is more suitable for using the design science research methodology. Unfortunately, there are still many researchers who do not understand or even do not know this methodology. In this paper, we discuss the design science research methodology framework to reintroduce this methodology. The discussion starts with the theory of design science research, which explains five stages, from the explicated problem to the evaluation stage. Furthermore, an example of the application of this methodology is presented for the timetabling problem. This application example shows how the design process was and what the results were. This paper is expected to provide knowledge to researchers, especially for researchers who should use this methodology.

This issue is the first of ISJ's 25th volume and, as is customary on such anniversaries, this editorial reflects on the life of the ISJ and on its future. The editors-in-chiefs have periodically analysed the nature of the papers that ISJ publishes, most recently in the valedictory editorial by David Avison and Guy Fitzgerald in volume 22 (Avison & Fitzgerald, 2012). This looked at papers in the first 21 volumes. Here we update this analysis by reflecting on the content in the last three volumes and consider how we need to develop the journal further. The last three volumes, 22-24, comprised 64 papers. ISJ has moved from four issues a year to six and the number of papers published in each volume has increased from around 17.3 to 21.3. The move to six issues a year was occasioned by increases in good quality submissions and a desire to ensure that the time between acceptance and publication was shorter. In these 64 papers, we see many of the publication trends identified in earlier papers exacerbated. The 64 papers were produced by 177 authors (some authors produced more than one paper, of course) with an average of 2.77 authors per paper. Remarkably, only 4 single authored papers were published (6.25%) cementing a decreasing trend of 32.4% single authored papers in volumes 1-17 and 13.6% in volumes 18-21. Papers with four authors are now quite common, and some have five or six. While single authored papers are often preferred by tenure and promotions committees, the trend in IS, or at least in the ISJ, is for multi-authored papers. ISJ is continuing to becoming more global, with the authors of papers more geographically diverse. Avison and Fitzgerald identified the reduction in UK papers in the ISJ, the plateauing of contributions from North America and the increases from Asia-Pacific. UK contributions in the most recent volumes have fallen from 16% to 14.2%. North American contributions drop from 40.8% to 36.7%, while Asia-Pacific authors now comprise 20.3% of the total (up from 18.9%) and Europe (excluding the UK) hosts 28.2% of contributors (up from 22.3%). Taken together, the increase in author numbers and their greater geographical spread means that more papers are co-authored by people from different geographical regions. Indeed, within regions there is a sense that co-authors work in different countries, and even within countries, co-authors are seldom co-located in the same institution – and where they are, they are often not in the same academic unit. Given the global nature of IS, and the multiple perspectives upon it that need to be brought to bear, this diversity appears healthy. In a different aspect of diversity, the gender balance of authors continues to become more even. 35% of the most recent paper authors are female, compared to 14.4% in volumes 1-6, 19.8% in volumes 13-17 and 28% in the volumes 18-21. As in the earlier analyses, virtually all the most recent authors are academics. Fewer than 4% of papers are co-authored by practitioners, down from 7.1% in the last analysis. While, we are improving diversity in many respects, we are not attracting (or publishing) papers by practitioners. If this reflects less research, or poorer research, being undertaken by or with practitioner co-authors then this is a concern for a discipline such as IS. The trend over time has been for more ISJ authors to be located in business and management schools or departments. Most recently, this percentage has increased again – from 43.3% to 46.3%. The percentage in IS departments has dropped again to 31.6%, while that from computer science has increased from a very low base to just over 5% and the number from ‘other’ has nearly doubled to 13%. However, caution needs to be exercised over the home academic units as the data is taken from the addresses of authors. Some authors may be in a department, within a school that is part of a faculty that is itself an element of a college which is a unit of an institution. Thus, someone may be in an IS department that is within a business school. The address they choose to use may refer to either (or both) – but perhaps reflects their preference for the unit they would most like to be associated with. Likewise, there is a diversity of titles for units that make classification problematic – departments of Tourism Information, Health Administration, Social Informatics, and Business Education for example. So, what does the ‘typical’ authorship of an ISJ paper look like? There will be multiple authors, of both genders who are unlikely to work in the same institution or even in the same country and whose departmental home disciplines are different. What these co-authors are researching remains diverse too. While ISJ has been known as an interpretivist journal around a third of papers are positivist. About half the papers are qualitative. In terms of the 10 questions and challenges that Avison & Fitzgerald (2012) posed in their final editorial, the ISJ has, recently, published more papers from under-represented regions, it has improved the gender balance of authors, and it has remained generalist in its outlook. We have published more papers, more issues and more special issues. We can continue to do better in these aspects, and other challenges remain. We would like to be more risk taking, we have plans for additional sections to reflect different aspects of IS and of research, and the journal needs to reflect the need that the IS community has identified for the impact of its research to be better understood and better communicated. We will take up these issues throughout this year. When we assumed the role of editors-in-chief in July 2012, we introduced a new era for the Information Systems Journal as management of the journal shifted to us from the founding editors. In our inaugural editorial we indicated our desire to retain the commitment to publishing high-quality research that is consistent with the founding mission of the journal. That is, the motivation for founding ISJ was a desire to publish work that pushed out the boundaries of acceptable methodology for conducting information systems research. Since the methodological norm in 1991 was quantitative and positivist research, ISJ was established to be a publication outlet for qualitative interpretive and later critical research. Going forward into the next 25 years we are committed to the founding principles of ISJ, as we interpret them today. First, we will continue to publish high quality research that is informed by the interpretive tradition. This means that while positivist and quantitative research is welcome, and indeed, has been published in significant quantity, the production of numbers and the testing of hypotheses are not sufficient. The paper must also ‘tell a good story’ with the literature review, the analysis of data, and the discussion. Second, we will continue to publish work that pushes out the boundaries of acceptable information systems research. But in 2015 this is less about methodology than it is about content. Therefore, while we continue to welcome and publish research about information system development and management, we also welcome and want to publish research on emerging topics of relevance to the information systems field. This would include, for example: papers about societal/environmental influences on information systems; research about information systems professionals; and the consequences of information systems – intended and unintended, positive and negative. We also encourage research on the effect of new technologies – such as 3D printing, ubiquitous computing, wearable computing, and the Internet of things – on information systems development and practice. Finally, we welcome research that will explore a future in which information systems are an ever more important component. For instance, information systems are increasingly embedded in organisations, in societies and even in people. However, these embedded systems and their implications for privacy, convenience or control are rarely subjected to in-depth investigations. Throughout 2015, as part of the ISJ 25th anniversary, we will publish, in addition to our regular papers and special issues, a set of invited papers. We invited all those who published in our very first volume, and some others, to consider submitting a paper for volume 25. Many accepted our invitation. It is remarkable how many of the people who published in ISJ 1(1-4) are still active in the field. This may reflect the youth of the discipline, the youth of those working in it 25 years ago, that ISJ attracted submissions from very young academics, or that continual reductions in pension entitlements globally have meant many people have continued working for longer than they had anticipated. The first paper in this issue is by Lee, Thomas and Baskerville who argue for the restoration of the IS artifact in design science. They do so by unpacking the concept of the IT artifact. In doing so they remind us that what is being designed is not the IT artifact, but the IS artifact, which consists of three artifacts: the technology artifact (the hardware and software), the information artifact (the message) and the social artifact (human actions). They illustrate the primacy of the IS artifact in design science with material from three cases. The result is to restore the idea that the study of design in information systems needs to consider the design of the entire IS artifact, not just the technology artifact. This result encourages an expansion in the use of design science research methodology to study a broader set of artifacts. Three cases serve as a reminder that the component parts of the IS artifact - the information artifact, the technology artifact, and the social artifact - all interact and combine into a whole that is greater than the sum of its parts. In the second paper, Myers and Baskerville contribute new methodological insights in their argument for active engagement by IS researchers during the conduct of ethnographic work. While most ethnographic research in information systems has been based on the traditional anthropological model in which the ethnographer observes but does not actively seek to change the situation, they argue for the incorporation of action research perspectives into ethnographic research. They do so by describing two existing types of research that relate ethnography and design. The first type is ethnography for design, in which the results of the ethnographic study of some context informs the subsequent design process. The second type is ethnography to study design. Here, the ethnographer is studying those who are engaged in the design activity. They contrast these with the third type, design ethnography, in which the ethnographer is actively engaged in the activity of design, production or introduction of artifacts. Hence, the researcher is intervening in changing the context being researched. The researcher is actively engaged with others in a future-oriented way: designing, creating, innovating, and improvising artifacts that may affect the cultural and social values under study. Hence, rather than being a passive observer of changes in human behavior or organizations and human society more generally, the design ethnographer becomes an active participant in shaping these changes. In the third paper of this issue, Alter passionately engages with the concept of the “IT artifact” and trenchantly argues that it is no longer useful, it has outlived its usefulness. IT artifacts are ubiquitous in research, yet there is a huge amount of inconsistency in the way the term is used, with the result that it no longer means anything in particular at all. Alter traces the way the definition of IT artifact has morphed since it was first used by Orlikowski & Iacono (2001) to definitions in 2013 that are so loose and all-encompassing as to be essentially meaningless. Alter also suggests that cousins of the IT artifact, notably the IS artifact, the sociotechnical artifact, the social artifact and the technical artifact, simply confuse the issue further, each overlaps to some extent with other artifacts. In order to fill the yawning void that will be created when the IT artifact is retired, Alter proposes three ways forward, the most ambitious of which develops Alter's own notion of IT-enabled work systems (Alter, 2013). As Alter notes in his conclusion, retiring the IT artifact will help to enhance the rigour of our research because we will not use terminology that is meaningless. Indeed, we will also focus on more useful research that is meaningful.

Based on reviewing foremost literature, this paper discusses various design science research methodologies and related to them case studies. I have identified common activities that may create an outline of a reference model for design science. The model will apply to a design step activity in any design science methodology. I also present my assumptions, based on structured interviews, that the current abstract level of design science methodologies may cause poor quality of content of its artefacts. As for the measurement of quality, I discuss representational information quality dimensions. The contribution of the paper is to relate design sciences theories with research activities by providing a design science process oriented framework. My observations indicate that it will help improve the quality of design science artefacts.KeywordsDesign Science MethodologySystematic FrameworkDesign Science Literature ReviewInformation Quality Dimensions

Knowledge-intensive business services (KIBS) such as consulting and research and development services are important factors of performance and innovation in industrialized economies. However, current modeling techniques aimed at supporting service design do not account for their core characteristics such as the relational nature of exchanges among providers, clients, and other actors. Using data from a case of academic research and development service as a type of KIBS, we present a modeling technique that can support the design of successful service engagements in this domain. This work is guided by the understanding of service as a process of collaborative value creation, or value cocreation. Beyond the contribution of the modeling technique to KIBS design, our work shows the strength of using a Design Science Research methodology in creating design artifacts that are strongly aligned with the problem domain for which they are developed.

This proposal outlines the design and development of a Java-based software artifact implementing a custom API Gateway to manage security and access control, replacing the need for JWT token systems. The proposal encompasses the definition of functional requirements, software design and architecture, implementation, and system deployment, aiming to offer a secure and efficient solution for communication and control between clients and a REST API. The Design Science Research (DSR) methodology was applied, focusing on the design and implementation process of the software product, which enables secure integration between a client and a REST API. Security in web applications and REST API architectures is essential to protect system integrity, confidentiality, and availability in an increasingly threat-exposed digital environment. Additionally, the Scrum methodology facilitates agile and collaborative project management, ideal for iteratively and adaptively implementing security solutions. During the design and coding phases in Java, the custom API Gateway is defined to manage incoming requests and safely route traffic to REST APIs. The custom API Gateway incorporates features such as authentication, user role validation, and request logging. Simple endpoints are created on the REST API to ensure seamless communication with the gateway. This proposal provides a robust solution for managing security and access control in distributed applications. By using the API Gateway instead of JWT tokens, the system centralizes security and authentication efficiently, enhancing fault tolerance and maintaining stable performance.

Government institutions collect and produce an extraordinary number of datasets to conduct and execute their programs and agendas. Various types of datasets collected by the governments can increase transparency and accountability, improve citizen engagement, and create value-added services for the public. Through the Open Government Data (OGD) initiatives, Non-Government Organisations (NGOs), private agencies, business enablers, data analysts, researchers, civil societies, and other open data stakeholders can take advantage of disclosing the government datasets. Despite its significance, the decision-making process to disclose government datasets is given limited attention and encounters several challenges. Although numerous datasets have been published to the public, many datasets remain undisclosed. Government institutions face several challenges in deciding to open datasets. First, the governments have not systematically analysed datasets to identify the benefits and disadvantages of opening datasets. Decision-makers, policy-makers, civil servants, and administrative officers do not know how to balance the advantages and disadvantages of opening datasets. Second, various stakeholders’ backgrounds may have different objectives and interests to analyse and disclose datasets. Third, the easy understanding of possible disadvantages of opening datasets results in moving away from the potential benefits due to the risk-avoiding culture in government. Therefore, these results in keeping datasets undisclosed. Furthermore, the stakeholders’ involvement in the decision-making process to open data, such as politicians, executive boards, decision-makers, civil servants, data analysts, and societies, all play essential roles and have different objectives for opening and using the datasets. For example, some decision-makers might have the authority to publish or keep the dataset closed. Some public servants might be risk-averse, whereas others might open datasets without considering possible negative consequences. As a result, the decision-making process becomes fuzzy, and the objectives of disclosing data are not reached. The different roles and interests of the heterogonous actors in the internal government organisation might create uncertainty and delay the decision-making process. Although there are guidelines, there are no decision-making tools to help governments decide to open their datasets. On the governments’ side, the potential disadvantages might easily dominate over the advantages. It is much easier for the decision-makers to keep a dataset closed than take the disadvantages of releasing a dataset. The lack of insights and expertise in estimating the potential advantages and disadvantages of opening data can also lead to uncertainty, which might result in avoiding the disclosure of datasets. Therefore, this research aims to develop Decision-making Support for Opening Government Data (DSOD). This DSOD accommodates a systematic approach to decide to open datasets. To achieve the objective of this research, we followed the Design Science Research (DSR) approach. The DSR approach results in developing a prototype of the DSOD as a design artefact and demonstrate it to the stakeholders.

PurposeThe study aims to explore and develop a smart route planning system for the cross-docking delivery operations of a large supermarket chain using an action research (AR) approach and assessing through a design science research (DSR) lens.Design/methodology/approachThis study took a problem-solving AR (PAR) approach toward the delivery operational issue of the case firm. The research process has accorded with the solution incubation and the refinement phases defined by a DSR framework. An intervention-based research framework for DSR is developed to assess the validity of this study as a DSR research and derive mid-range theories.FindingsDramatic operational and financial improvements were achieved for the case firm. Significant and unintended environmental and social benefits were also found. A design proposition (DP) and several mid-range theories are proposed as an extension of AR research to DSR research.Research limitations/implicationsA problem-solving DSR research can be better assessed by the intervention-based DSR framework developed in this study. DSR studies should be encouraged for both practical and theoretical advancement purposes.Practical implicationsA challenging business problem-solving study can be tackled effectively through an industry/academic collaboration taking a PAR approach to deliver substantial values and organization transformational results.Social implicationsDrivers and store associates are safer with smart delivery operations in the case firm.Originality/valueThere are still limited PAR design science case studies in the supply chain/logistics research literature. The research experience and findings gained from this study provide more insights toward how this type of research can be conducted and assessed.

The emerging impact of cloud computing in businesses has become a great driving force for business success and cannot be understated. The lack of a tailored functional application software, even though a framework exists, poses a significant challenge to SMEs. Without a digital corresponding software solution, SMES struggle to conduct a practical assessment since the framework offers only theoretical guidance. A well-developed cloud implementation assessment tool is essential for SMEs to thrive. The main aim of the study was to develop a robust cloud implementation assessment system for SMEs to evaluate their cloud implementation readiness. Design Science Research Methodology (DSRM), is a kind of research methodology which aims at developing and evaluating innovative artefacts, in the form of models, frameworks, designs, or prototypes, to solve complex and practical problems. Framework Usability Validation used. On the Clarity of the framework, none of the respondents strongly disagreed, representing 0%. 8 respondents, representing 15.4% disagreed, and 3 respondents, representing 5.8% were unsure. The system was developed using the Django Python web framework. Out of 43 respondents, 18 participants (42%) found the Learn Cloud Phase helpful, 3 participants (7%) found the Assessment and Cloud Choice Phase helpful, and 15 participants (35%) found the Migration Phase helpful. Additionally, 4 participants (9%) reported that all phases were helpful, 2 participants (5%) indicated that none of the phases were helpful, and 1 participant (2%) was uncertain about the helpfulness of any phase. In this paper, the design science Research (DSR) method was utilised to develop and evaluate the cloud implementation assessment system for SMEs. Key features of the developed system include learn cloud, assessment, cloud choice and migration phase. The developed system was said to be very useful and suitable, solidifying its value as a helpful resource for small businesses aiming to utilise cloud technology. A cloud assessment system was developed to aid in assessing SMEs' cloud readiness and implementation. Generally, SMEs responded positively to the cloud computing implementation framework.

Correctly identifying and classifying food is decisive in food safety. The food sector is constantly evolving, and one of the technologies that stands out is augmented reality (AR). During practical studies at Companhia de Entreposto e Armazéns Gerais de São Paulo (CEAGESP), responsible for the largest food storage in South America, difficulties were identified in classifying aromatic herbs due to the large number of species. The project aimed to create an innovative AR application called ARomaticLens to solve the challenges associated with identifying and classifying aromatic herbs using the design science research (DSR) methodology. The research was divided into five stages according to the DSR methodology, from surveying the problem situation at CEAGESP to validating the application through practical tests and an experience questionnaire carried out by CEAGESP specialists. The result of the study presented 100% accuracy in identifying the 18 types of aromatic herbs studied when associated with the application’s local database without the use of an Internet connection, in addition to a score of 8 on a scale of 0 to 10 in terms of the usability of the interface as rated by users. The advantage of the applied method is that the app can be used offline.

PurposeThe main goal of this research work is the optimization of the purchasing business process in the Moroccan public sector in terms of transparency and budgetary optimization. The authors have worked on the public university as an implementation field.Design/methodology/approachThe design of the research work followed the design science research (DSR) methodology for information systems. DSR is a research paradigm wherein a designer answers questions relevant to human problems through the creation of innovative artifacts, thereby contributing new knowledge to the body of scientific evidence. The authors have adopted a techno-functional approach. The technical part consists of the development of an intelligent recommendation system that supports the choice of optimal information technology (IT) equipment for decision-makers. This intelligent recommendation system relies on a set of functional and business concepts, namely the Moroccan normative laws and Control Objectives for Information and Related Technology's (COBIT) guidelines in information system governance.FindingsThe modeling of business processes in public universities is established using business process model and notation (BPMN) in accordance with official regulations. The set of BPMN models constitute a powerful repository not only for business process execution but also for further optimization. Governance generally aims to reduce budgetary wastes, and the authors' recommendation system demonstrates a technical and methodological approach enabling this feature. Implementation of artificial intelligence techniques can bring great value in terms of transparency and fluidity in purchasing business process execution.Research limitations/implicationsBusiness limitations: First, the proposed system was modeled to handle one type products, which are computer-related equipment. Hence, the authors intend to extend the model to other types of products in future works. Conversely, the system proposes optimal purchasing order and assumes that decision makers will rely on this optimal purchasing order to choose between offers. In fact, as a perspective, the authors plan to work on a complete automation of the workflow to also include vendor selection and offer validation. Technical limitations: Natural language processing (NLP) is a widely used sentiment analysis (SA) technique that enabled the authors to validate the proposed system. Even working on samples of datasets, the authors noticed NLP dependency on huge computing power. The authors intend to experiment with learning and knowledge-based SA and assess the' computing power consumption and accuracy of the analysis compared to NLP. Another technical limitation is related to the web scraping technique; in fact, the users' reviews are crucial for the authors' system. To guarantee timeliness and reliable reviews, the system has to look automatically in websites, which confront the authors with the limitations of the web scraping like the permanent changing of website structure and scraping restrictions.Practical implicationsThe modeling of business processes in public universities is established using BPMN in accordance with official regulations. The set of BPMN models constitute a powerful repository not only for business process execution but also for further optimization. Governance generally aims to reduce budgetary wastes, and the authors' recommendation system demonstrates a technical and methodological approach enabling this feature.Originality/valueThe adopted techno-functional approach enabled the authors to bring information system governance from a highly abstract level to a practical implementation where the theoretical best practices and guidelines are transformed to a tangible application.

Physicians working in emergency departments (ED) face significant challenges due to inadequate clinical decision support and the fragmentation of hospital information systems (HIS). This study develops a system integration architecture that facilitates the creation of a clinical decision support system (CDSS) based on machine learning to improve decision-making in the ED. Design science research methodology was employed to create and evaluate the designed architecture which provides integration of HIS systems within the ED setting. The research enables seamless accessing and processing of data from disparate HIS systems, allowing the implementation of machine learning techniques for enhanced clinical decision support. The designed system integration architecture enables the CDSS and allows physicians to address patients' data more effectively, leading to better-informed clinical decisions and significantly enhancing decision support capabilities in the ED, contributing to improved healthcare outcomes and hospital expenditure by leveraging machine learning techniques.

Design Science Research (DSR) is a paradigm that centres on the development and evaluation of artefacts. While central, a proper scientific evaluation can significantly increase the work required to complete a DSR project. The high costs, workload, and delays caused by evaluations have been shown to lead to poor or absent evaluations. These findings present two paradoxes: (1) evaluation is considered essential, yet it is commonly done poorly or not at all; (2) DSR promises to improve research relevance through timely artefact development and reliability through scientific evaluation, yet poor and lengthy evaluations cause DSR to often fail to deliver on its promise. To address these paradoxes, this paper presents MEDS (Methods for Evaluation in Design Science), a four-step evaluation method that includes three component methods new to the evaluation guidance literature: MuSCoW (Must, Should, Could, Won’t), the DSR Evaluation Selection Framework, and Short-Scoping of evaluation methods. MEDS aims to address the above paradoxes and guide DSR researchers, especially novices, to achieve a programme of effective and resource-efficient DSR evaluations as an on-going component of existing DSR methodologies. By carefully planning, orchestrating, and scoping evaluation activities, MEDS delivers a rigorous evaluation programme without overwhelming researchers with a massive evaluation burden.

With the advent of machine learning (ML) techniques, various algorithms have been applied in previous studies to develop models for predicting soil fertility status. However, these models are observed to use varying fertility target classes, and variations have been reported in these models' predictive performances. As a result, practical applications of these models for obtaining the most accurate predictions may become hindered. While the weighted voting ensemble (WVE) ML technique can be used to improve soil fertility status prediction by aggregating individual models prediction, guaranteeing finding of an optimal WVE assignment weights is challenging. Whereas a brute exhaustive search procedure can be applied for the mentioned task, there is a lack of exploration on the exploitation of automated classifiers' precise weights combinations as search spaces for successful optimization. This research aims to develop a high-performance soil fertility status prediction voting ensemble using brute exhaustive optimization in automated 1EXP(-)Z+ multi-precision weights of hybrid classifiers. Soil chemical properties and ML modeling algorithms for modeling soil fertility status were identified. Base hybrid ML classification models for predicting soil fertility status were evaluated using Tanzania as a case study. Finally, the base ML hybrids WVE models were optimized using brute exhaustive search procedure’s novel developed search spaces generation algorithm for guaranteed optimal solution finding. The research was designed using design science research methodology, with the application of unsupervised machine learning K-mean algorithm with a knee detection method to find the optimal number of soil fertility status target classes, and supervised learning algorithms were applied to model classifiers for those optimal classes. Three soil fertility target classes were identified by clustering technique. The model achieved on test data a predictive accuracy of 98.93%, with respective AUC of 82%, 83%, and 87% for low, medium, and high soil fertility targets classes. Whereas these performances are observed higher compared to models in previous studies, 92% correct classifications were obtained on validation against external unseen laboratory-based tested soil results. Therefore, soil testing laboratories and farmers should consider using the model to smartly manage soil fertility which may lead to improved crop growth and productivity. The government could set agricultural-related policies that require the use of the model by farmers with the provision of agricultural inputs subsidies. Future work could be to develop an integrated real-time web and mobile application for providing farmers with soil fertility status information.