Abstract: In this digital era, data is the new oil and artificial intelligence (AI) is the new electricity, which is needed in different elements of operations management (OM) such as manufacturing, product development, services and supply chain. This study explores the feasibility of AI utilization within an organization on six factors such as job-fit, complexity, long-term consequences, affect towards use, social factors and facilitating conditions for different elements of OM by mining the collective intelligence of experts on Twitter and through academic literature. The study provides guidelines for managers for AI applications in different components of OM and concludes by presenting the limitations of the study along with future research directions.

Grover, P., Kar, A.K. & Dwivedi, Y.K. Understanding artificial intelligence adoption in operations management: insights from the review of academic literature and social media discussions. Ann Oper Res (2020). https://doi.org/10.1007/s10479-020-03683-9

This article explores the utilization of the AI in different elements of the OM, manufacturing, product development, services and supply chain through academic literature and expert (workers, professionals, corporates & organizations) opinion on Twitter. Through the study, authors attempted to expand the knowledge of managers and organization on AI utilization within an enterprise by combining academic literature and practice driven discussions in social media.

A systematic literature review has been undertaken in academic literature to address the propositions developed for our study. The though behind undertaking the systematic literature review is that it will capture the state of literature that is well developed and established with fair degree of accuracy at the cost of recency. To address recency, data collection has been done in Twitter where extraction step in Twitter is followed by data pre-processing step. The pre-processing sub-steps includes data transformation, storing of data, data reduction and data cleaning. For data analysis, descriptive analysis and text analytics had been used. Descriptive analysis provides the first level insights into the data by applying the simple functions such as count, sum, mean, median and mode. Text analytics extracts information from textual data. The collective complementary intelligence of different stakeholders, namely academia and practitioners, have been used in this study to build and validate propositions by taking inputs from complementary sources and validating them through an triangulation like approach, conceptual, practical and innovation.

Now we attempt to summarize the insights extracted from academic literature and social media analytics on Propositions 1a, 1b, 2a, 2b, 3a, 3b, 4a and 4b. AI utilization factors had been framed by getting inspiration from model of PC utilization constructs, job-fit, complexity, long-term consequences, affect towards use, social factors and facilitating conditions.

For Proposition 1a academic literature gives a positive signal on usage of AI for big size product inspection. Usage of AI will enhance the product quality. Academic literature indicates usage of AI as quality deployment function will be beneficial for large scale production. Especially when humans get exhausted, then AI powered system will play a great role in maintaining product quality standards, which is in line with literature which indicates AI usage for inspection reduces cost and time spent on the inspection. On social media experts were discussing on service robots and AI powered drone usage for inspecting steel structures and wind turbine respectively. For Proposition 1b exact use cases were not available but there were tweets in which usage of software’s XRC and NDT had been discussed for creating professional inspection reports.

Academic literature indicates AI implementation on ERP is best suited for MNCs working across the world, which may lead to minimization of resources and improving overall productivity. The biggest hurdle in implementation is selection of data sources or to apply on which databases or data warehouses maintained by organizations. Explainability of difference of outcome to internal employees may also be a challenge. Another hurdle in implementation is that AI algorithms for big data, characterized by high volume, variety, veracity and velocity, have to demonstrate maturity. On social media, experts have positive opinion on AI algorithm usage for learning and identifying core capabilities of an organization. Some experts had highlighted AI utilization with an organization will make the organization “smart” as well as hyper productive. This will subsequently will help the organization in developing digital business and innovation.

Literature indicates AI implementation for internal and external stakeholders is ideal for those mature organization who really wants to capture the large segments of the market. Appropriate suggestion by the AI algorithm for each customer is the challenging job for the recommender system. Academic literature contains positive as well as negative examples for using AI powered system for customer relationship management. On Twitter, experts did not have significant differences among positive and negative opinion regarding the use of AI for connecting with external or internal stakeholders. Experts on Twitter had indicated natural language processing and deep learning will facilitates the AI implementation for  enhancing customer experiences.

Literature indicates implementation for upstream supply chain will be demanding in the situations where budget is under consideration. AI implementation will lead to reduction in procurement costs by providing employees with diverse and different options. For downstream supply chain, AI implementation may be favourable when organizations are launching new products. AI will help the organizations in elicitation of requirements and as well as predicting what will be costumer’s preference among requirements in future and may help in new product development. Employee and customer rigidity is the biggest hindrance in AI implementation in the upstream and downstream activities, which is in line with literature finding. Experts on Twitter holds positive opinion on implementing AI in upstream and downstream supply chain management within an enterprise. They reveal that such an AI implementation within supply chain can be augmented through the use of blockchain.