To Use or Not to Use Artificial Intelligence? A Framework for the Ideation and Evaluation of Problems to Be Solved with Artificial Intelligence

Abstract

The recent advent of artificial intelligence (AI) solutions that surpass humans' problem-solving capabilities has uncovered AIs' great potential to act as new type of problem solvers. Despite decades of analysis, research on organizational problem solving has commonly assumed that the problem solver is essentially human. Yet, it remains unclear how existing knowledge on human problem solving translates to a context with problem-solving machines. To take a first step to better understand this novel context, we conducted a qualitative study with 24 experts to explore the process of problem finding that forms the essential first step in problem-solving activities and aims at uncovering reasonable problems to be solved. With our study, we synthesize emerged procedural artifacts and key factors to propose a framework for problem finding in AI solver contexts. Our findings enable future research on human-machine problem solving and offer practitioners helpful guidance on identifying and managing reasonable AI initiatives. 2.1. The Process of Problem Solving Problem solving, the act of uncovering problems and searching for effective solutions [20, 53, 54, 56], is

How to facilitate manufacturing industry learning from problems: a review on advanced technology problem-solving

“…As a countermeasure to the identified Mode-1 conditions on the subject, the argument here is that contemporary research on the subject, and in extension, the practical adoption of such results, should embrace the central ideas of Mode-2 problem-solving and learning in advanced technology implementations for problem-solving purposes. This infers taking the time to go to the problem, as advocated by Sturm et al (2021), reflect on it properly (Shahlaei and Snis, 2022), test existing assumptions about it, expand the knowledge about it, and hence, expand the knowledge on potential solutions as well as choosing sustainable courses of action (Argyris and Schön, 1997; March, 1991; Mohaghegh and Furlan, 2020; Watzlawick et al , 2011). Although consciously adopting Mode-2 may result in excessive abstractedness (Liedtka, 2020) and a loss of using existing capabilities (Mohaghegh and Größler, 2021), its explorative nature may on the other hand contribute to the generation of new knowledge about potential use of advanced technology implementations for problem-solving.…”

“…This is grounded in the scoping review findings showing a lack of empirical anchoring of tested algorithms and models (see Table 1), as well as the identified solution focus of the field. By including more empirically anchored studies, both the initial and implementation phases of problem-solving processes could be further understood, as advocated by Sturm et al (2021) and Joksimovic et al (2023). Further research on empirically anchored applications may also support manufacturing industries in understanding added value of advanced technology, as commented by Burggräf et al (2021).…”

“…A recent review by Joksimovic et al (2023) on the application of Artificial Intelligence (AI) in problem-solving activities mapped contemporary research on the subject and found that many studies are indeed found in the areas of engineering and industry; however, both theory development and implementation focused studies are not prioritized. It has also been found that research connected to Industry 4.0 is most often focused on the actual technologies and optimization problems (Zheng et al , 2021), and research on advanced technology in general tends to have a strong solution focus (Sturm et al , 2021). Despite these focuses in research, companies face a lot of other problems with different characteristics daily, including problems related to worker skills, inventory, supply chain disruptions, worker safety, consumer trends, maintenance and innovation.…”

Artificial Intelligence Augmented Project Management

“…According to research by IDC 4 , 40% of Digital Transformation will be related to AI, while Accenture 5 predicts that 80% of finance accounting tasks will be automated in the upcoming years.…”

“…At the present moment, there is no widely recognized model, structure, or standard for coming up with AI applications in Project Management. A researcher has developed a new method to identify use cases given the technology and business problem [4] but not specifically for the Project Management domain. As of today, AI application in Project Management can be seen as still in the infancy stages.…”