Sociotechnical interdependencies and tipping‐point dynamics in data‐intensive services

Abstract

Service science theories do not fully explain failure cases in data‐intensive services – high‐technology services that utilize large volumes of data and provide customized information for users. In these service systems, the technological and social elements are highly interconnected: firms cannot maintain databases and analytic capabilities if they lack market penetration, which itself is influenced by performance of analytic capabilities and databases; a positive feedback loop. Informed by a case study, we develop a simulation model, postulate that market adoption is highly sensitive to the launch conditions, specifically with respect to the initial states of the system, and offer a dynamic theory in which highly nonlinear relationships among the initial states of the system drive outcomes of the market adoption of data‐intensive services. We argue that a tipping point exists at which small technological differences in the launch period differentiates between market success and failure. Furthermore, our analysis points to a strong self‐fulfilling mechanism whereby initial positive customer perception can increase the service's likelihood of success by indirectly influencing technological improvement. In contrast to the conventional belief, having a large number of initial adopters may negatively influence market adoption in the long run. A major source of the patterns observed is the high level of interdependencies between social and technological elements of data‐intensive services. © 2022 System Dynamics Society.

Generative agent‐based modeling: an introduction and tutorial

“…Fifth, the specific example used in this paper also brings insights into the body of system dynamics modeling literature that deals with dynamic diffusion models. Such models include a wide range of market diffusion (Barabba et al ., 2002; Ghaffarzadegan et al ., 2023), norm diffusion (Ulli‐Beer et al ., 2010), organizational change (Wunderlich et al ., 2014) and epidemic models (Darabi and Hosseinichimeh, 2020; Rahmandad, 2022; Rahmandad and Sterman, 2022). It also provides a case of path dependence behavior, a problem commonly studied in dynamic models (Sterman, 2000).…”

Preparing for pandemic response in the context of limited resources

“…However, as the system dynamics tradition has long emphasised, resources are stock variables that accumulate over time at different speeds (Ghaffarzadegan et al ., 2023). From a resource‐based viewpoint (Barney, 1991; Peteraf, 1993), variation in performance relates to the differences in the rate at which organisations build and maintain their capacity.…”

“…We frame this work as a case‐based modelling approach (Ghaffarzadegan et al ., 2023), drawing upon knowledge gained from past crises. This knowledge has emerged as a result of a project which brings together expertise from a number of European national public health agencies, aiming to enhance capacity for pandemic preparedness and response.…”