A saddle or slide instead of a bell: Agent-based modelling and simulation of unusual curves for the adoption or diffusion of innovations

• Applicant: Professor Dr. Christian Stummer
• Subject Area: Management and Marketing
• Term: since 2025
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 562433469

Research on the market diffusion of innovations has a long tradition. While initially analytical models were used for this purpose - with the Bass model originating from 1969 as the most prominent one and starting point for numerous extensions - a new family of approaches have become increasingly established since the turn of the millennium. These new approaches use agent-based modelling and simulation (ABMS) to analyze complex market behavior when new products are introduced. In such models, the stakeholders (e.g., the consumers) are represented as independent "agents", who can have heterogeneous preferences, interact with each other in a social network, make individual experiences over time, and can later-on decide for or against the adoption of the innovation (e.g., the purchase of the new product). In many cases, these ABMS approaches model the adoption process and, thus, innovation diffusion more realistically than traditional approaches. In nearly all models from both families - that is, analytical as well as ABMS-based innovation diffusion models - a bell-shaped adoption curve or a corresponding S-shaped diffusion curve is assumed. However, empirical studies indicate that adoption curves can show the form of a saddle or a slide instead of a bell. In our research proposal, we intend to set up an agent-based model and implement it into a simulation tool with which the occurrence of such unusual curves can be analyzed. We will systematically analyze market characteristics for which such curves occur, and we plan to use data from diffusion curves from real products that show the properties of a saddle or slide curve to demonstrate that our approach can nicely reproduce these curves. Regarding the time needed as well as the funds requested, our project proposal is relatively lean, which is possible due to experience from earlier projects. Despite the small scope of the project, we will still be able to investigate an empirical phenomenon (namely, the occurrence of unusual curves) with a method not yet used for this particular purpose (namely, ABMS) by varying both the characteristics of the stakeholders and the social network, which is not possible in a comparable way by means of traditional analytical approaches. The theoretical contribution of this project lies in the systematic analysis of the impact of market characteristics on the occurrence of unusual adoption and diffusion curves. The corresponding findings can support practitioners in planning countermeasures for possible saddle curves. The methodological contribution is the demonstration of the applicability of an ABMS approach in tackling this type of research question.

• DFG Programme: Research Grants
• International Connection: Sweden
• Cooperation Partner: Professorin Dr. Franziska Klügl