Information Systems Research Methodology for the 2030'es.
HICSS-59 Mini-Track in the Decision Analytics and Service Science Track
Keywords: Digital fusion for decision support, Digital coaching, Explainable algorithms, Fast online problem solving, AI and machine learning
CALL FOR PAPERS
The prevailing ISR methodology has been built on decision analysis and design science for the creation, development and use of decision support systems, which have been major contributions to beter decisions and enhanced productivity in (e.g.) transportation, finance, health care, education, cyber security, etc. The common denominator in most research inquiries is the accumulation of human cognitive power by using as support the analytic capabilities of computers.
The rapid development of artificial and computational intelligence is challenging the assumption that only human cognitive power matters as these technologies can learn and build new cognitive skills. Decision makers still carry full responsibility for the decisions and the consequences they bring. Decision analytics was developed as a partial response to the emergence of intelligent, analytic technologies. It is now clear that such pursuit is not enough: new principles and guidelines are needed for ISR methodologies in the 2030'es as the cognitive capabilities of AI technologies grows. The new principles and guidelines will evolve from and build on new forms and levels of cognitive power to accumulate innovative elements of decision analytics, artificial and computational intelligence.
This mini track invites papers on expected novel and speculative methodological forms, methods and elements that can serve as foundational elements of new methodologies for IS research. The papers can build on descriptions and evaluations of ongoing business innovations, assessments of industrial best practises and corporate cases that introduce novel problems. The papers will show management problem-solving and decision approaches that could be better and more productively handled if they were guided by innovative and enhanced ISR methodologies and related research results which are adaptive to the new organizational challenges and tasks.
The current principles of decision analysis (DA) have been reinterpreted, enhanced, and adjusted over 3-4 decades to meet the needs from growing complexities of large, multinational, dynamically interdependent industries, and corporations that in ever growing competition adapt to dynamically evolving innovations. The reinterpretations have formed foundation for operational research, management science, multiple criteria decision making, etc. and have offered formal frameworks for important decisions from the 1980'es through the 2010'es. The DA theoreticians are, however, not overwhelmed by the thousands of success stories … "while there may occasionally be justification for such methods in applications, decision analysts would argue for multi-attribute utility theory as the gold standard to which other methods should be compared, based on its rigorous axiomatic basis". Lotfi Zadeh (in an HICSS keynote address) had some misgivings in the era of massive streaming big data and formulated it as "you can increase precision if you are willing to give up on relevance or you can increase relevance if you are willing to give up on precision, but you cannot do both at the same time".
Information systems research aims at using information technology to handle and solve organizational problems and improve or change organizational tasks and related human activities. Design science has become a key part of this process as it creates novel artifacts (algorithms, human/computer interfaces, process models, languages, etc.) which extend the boundaries of human capabilities to address problems and tasks that so far have not been tractable. The design-science paradigm has guided problem-solving with research informed innovations that define theoretical ideas, new practices and create visions through which analysis, design, implementation, management and use of information systems can be more effective and productive.
The Decision Support Systems (DSS) movement (first), methodology (rather soon after) and paradigm (gradually) started in the 1980'es. Managerial tasks are not routine, and managers asked for "support to do a better job". DSS methods were shown to offer both simpler and better guidance than methods developed within Decision Analysis – they offered service, fast delivery, ease of use, benefit focused more than cost, imprecision allowed for timely delivery and user control; these were messages that got support in management. DSS builders focused on the users' priorities, they developed systems linked to key business activities and they viewed the quality of a system from the value it gives to users rather than how advanced the offered systems technology was.
Decision support systems applied DA methodology standards for the first 2-3 decades for verification and validation of models that were used in decision support systems. This did not work out too well in several studies as DA models follow levels of abstraction that are not well suited for "support to do a better job". In the 1980'es the prevailing management science paradigm added challenges with "black box" optimization to find and develop better decisions. In case human cognitive ability was not enough, optimization algorithms took over ("replaced humans", if we like) and offered the best possible solutions. The optimization algorithms were beyond the professional knowledge and mathematical skills of the users who sometimes did not see why the presented optimal solutions would be the best possible way for any given problem situation.
Decision analytics represents a shift from both DA and DSS towards developing and delivering critical data, information and knowledge for management, decision-making, negotiations, planning, operations, (public, private sector) administration, etc. Analytics builds on theory and advanced algorithms as part of information systems. Research used for decision analytics in the 2020'es shows themes like big data, machine learning, business and service analytics, gamification, virtual and augmented reality, visual decision analytics, soft computing, computational logistics, explainable AI, etc. These all are described as "hot topics", which offer stronger results than DA and DSS, and that are relevant for contemporary, competitively effective management. ISR methodologies for the 2030'es will go beyond these and are expected to offer frameworks and conceptual models and approaches that will make full use of insights and contributions achieved with advances in decision analytics and AI.
An example of such trends is a joint industry and university research program (D21 in Finland). In this program the proposition was to build on human and system joint intelligence that can use fast, automatic algorithms with large, well-structured datasets and which combines this analysis with knowledge mobilized from seasoned experts. To make it work, system users need context relevant advice (in real time, with real data and information) adapted to their cognitive abilities and background knowledge, i.e., advice they can understand and use. This process is denoted as digital coaching, which combines algorithmic data with linguistic information and knowledge in automated joint intelligence systems.
Digital coaching requires that we master the transition from data to information, and on to knowledge, now known as digital fusion. Data fusion collects and harmonizes data from a variety of sources with different formats and labels. Information fusion uses analytics to build syntheses of data to describe, explain and predict key features for problem solving and decision-making. Knowledge fusion uses ontology to build and formalize insight from data and information fusion as a basis for computational intelligence methods, AI, machine learning, soft computing, approximate reasoning, etc.
A new ISR methodology will probably take form as a synthesis of the successful elements of decision analysis, design science, decision support systems and decision analytics but with additional elements (both cognitive and digital) from digital coaching, which probably will use digital platforms to make artificial and computational intelligence methods operational for practical problem-solving and decision making.
With this new mini track, we seek contributions from researchers with experience and interest in theoretical ISR issues and challenges, but also with experience of finding practical solutions for information systems. As the aim is for the 2030'es there will be innovations in technology that could resolve some of the challenges we have to deal with today. On the way to the 2030'es there is much that can be accomplished with the methods, technology, skills and experience of today to get "support to do a better job". We are interested in papers reporting on the "what and how" and why the users accept and adopt the methods. The context can be (but is not restricted to) the business decision support, scenario planning, digital health, optimal logistics planning, digital service economy, algorithmic and cognitive computing, digital coaching, digital services, and digital service systems management. We are interested in contributions where the applied/defined ISR methodologies become visible with either an experimental or empirical focus. Innovative studies working on new, emerging problems – that have become visible with new technology and new context descriptions - and based on explainable methods are most welcome. We look for contributions which combine innovative theoretical results with sufficient empirical verification, or good empirical problem solving, planning or decision making with new systematic theory building. A common denominator for all studies is the use of and contributions to elements, constructs and insights for a (possible) ISR methodology for the 2030'es. Topics that are appropriate for this mini track include, but are not limited, to:
• Acceptance and use of ISR methodologies for solving contemporary large problems
• Artificial intelligence and large language models in management
• Delegation and monitoring of AI during decision making
• Automation and augmentation in organizational decision making
• Risk associated with machine learning based decision making
• Explainable algorithmic and heuristic decision making for large problems
• Natural language processing for information and knowledge support
• Cognitive computing for design and management of digital services
• Soft computing for digital coaching
• Fuzzy logic and fuzzy decision making (precision vs. relevance issues)
• Digital fusion for decision support
• AI and machine learning for effective operational management
• Machine learning and fast online problem solving
• Deep learning approaches for handling large, complex planning problems
• Digital fusion of data, information and knowledge
• Digital coaching with support from automated joint computing systems
• Testing/evaluations of ISR methodology impacts on organizational performance
• ISR methodology enhancement with interactive visualization and visual analytics
We seek to to select high quality papers from this mini track for a Special Issue of a leading ISR journal, which will include an editorial on Information Systems Research Methodology for the 2030'es. The Special Issue will be followed by 2-3 more collections of papers from HICSS-60/61 as the developments of the ISR new methodology grows. The selected papers for the first SI will need to be expanded to fit the requirements for standard research articles to be published in a leading ISR journal. The mini track co-chairs will guide the selected papers towards final publication, but further reviews may be needed in line with the editorial policy of the selected journal.
June 15: Paper Submission Deadline (11:59 pm HST) to the HICSS-59 website.
Minitrack Co-Chairs:
Christer Carlsson (Primary Contact)
IAMSR, Abo Akademi University
Yong Liu
Aalto University
Kalle Lyytinen
Case Western Reserve University
Jozsef Mezei
Abo Akademi University
------------------------------
CHRISTER CARLSSON
Professor
Institute for Advanced Management Systems Research
Turku
------------------------------