Exploratory analysis of the impact of information dynamics on innovation diffusion

Understanding the dynamic nature of innovation diffusion processes, and mechanisms underlying these dynamics is crucial, since such an understanding is potentially very important in designing effective innovation support policies and developing better diffusion forecasts. The role of information diffusion in conditioning the diffusion dynamics of an innovation is the locus of this study. In order to investigate this, a simulation model that distinguishes between the real attributes of the innovation and their perceived levels by the user groups has been developed. The model makes it possible to separately trace the diffusion dynamics of innovation and the information about an innovation. Additionally, the formulation of the model enables the message broadcasted via word-of-mouth to change in nature from positive to negative, or vice versa. This generic model is used in an exploratory way, which is discussed as a novel approach for conducting a simulation-based analysis. Such an exploration covers a wide range of plausible diffusion behaviors, and aims to demonstrate the extent to which information imperfections and dynamics may influence the diffusion process. During experiments it is observed that information imperfections as well as the pace of learning processes may yield significant changes in the diffusion patterns. These changes may be in the form of altering the basic characteristics of the well-known S-shaped diffusion curve, as well as stopping the diffusion at much lower levels than full adoption. The analysis presented in the article shows that exploratory analysis is a promising way to utilize simulation models for developing general insights about dynamics processes.     

Innovation diffusion with generation substitution effects

Markets of high technology products and services, such as telecommunications, are described by fast technological changes and rapid generational substitutions. Since the conventional modeling approaches that are based on diffusion models do not usually incorporate this important aspect into their formulations, the accuracy of the provided forecasts is consequently affected. The work presented in this paper is concerned with the development of a methodology for describing innovation diffusion, in the context of generation substitution. For this purpose, a dynamic diffusion model is developed and evaluated, based on the assumption that the saturation level of the market does not remain constant throughout the diffusion process but is affected by the diffusion of its descendant generation, as soon as the latter is introduced into the market. In contradiction to the conventional diffusion models, which assume static saturation levels, the proposed approach incorporates the effects of generation substitution and develops a diffusion model with a dynamic ceiling. The importance of such an approach is especially significant for markets characterized by rapid technological and generational changes. Evaluation of the proposed methodology was performed over 2G and 3G historical data and for a number of European countries, providing quite accurate estimation and forecasting results, along with important information regarding the rate of generation substitution.     

Forecasting and analyzing the competitive diffusion of mobile cellular broadband and fixed broadband in Taiwan with limited historical data

Taiwan experienced the rapid growth of mobile cellular broadband from 2005 by introducing 3G operations and had higher penetration than the average of the developing countries, the world, and even the developed countries. There are many forecasting models which were developed and successfully predicted the diffusion of long lifecycle product, but there are very few forecasting models which were developed for predicting new products with short lifecycle. Assumption of these models is always the growth of products follows an S-shaped curve. As for the products which were just introduced to the market, it is very difficult to identify if they follow an S-shaped curve with their limited historical data. This research aims to apply Grey system theory to predict the diffusion of mobile cellular broadband and fixed broadband in Taiwan since Grey system theory has a characteristic which requires very limited primitive data (the least 4 data) to build a differential forecasting model. We use penetration as an indicator to describe the diffusion of new products. The numerical data show that the Grey forecasting models GM(1,1) built in this paper have higher prediction accuracy than logistic models and grey Verhulst models. Moreover, we apply Lotka–Volterra model to analyze the competitive relationship between mobile cellular broadband and fixed broadband. The empirical data show that the relationship is commensalism rather than predator–prey. These results can be extended to contribute to other researches.     

Agent-based modeling of the diffusion of environmental innovations — An empirical approach

This paper presents an agent-based model of the diffusion of water-saving innovations. The empirical foundation of this model is a study, which was carried out for that specific purpose. As an example case, the diffusion of three water-related innovations in Southern Germany was chosen. The model represents a real geographic area and simulates the diffusion of showerheads, toilet flushes, and rain-harvesting systems. Agents are households of certain lifestyles, as represented by the Sinus-Milieus® from commercial marketing. Agents use two different kinds of decision rules to decide upon adoption or rejection of the modeled innovations: A cognitively demanding deliberate decision rule and a very simple decision heuristic. Thus, the model integrates concepts of bounded rationality. The overall framework for decision-making is the Theory of Planned Behavior, which has been elaborated using innovation characteristics from diffusion research. The model was calibrated with empirical data stemming from a questionnaire survey and validated against independent data. Scenarios for the nearer future show that water-saving innovations will diffuse even without further promotion, and different promotion strategies that relate specifically to both innovations and lifestyles can be pointed out.     

Innovation diffusion models expressing asymmetry and/or positively or negatively influencing forces

Two different characteristics of the innovation diffusion process, that is, the asymmetry and the appearance of positively or negatively influencing forces, are analyzed. Related diffusion models in use are presented and new generalized models are formulated. A five-parameter model is proposed in order to cover the above two different characteristics of innovation diffusion. The properties of this model and other related diffusion models are examined and the appropriate iterative nonlinear regression analysis technique is developed. Five comparative applications follow, predictions are made, as are comparisons between two generalized models—GRM₁ and Von Bertalanffy—that express asymmetric diffusion behavior.     

Binomial innovation diffusion models with dynamic potential adopter population

Models for describing the time pattern of the diffusion processes for innovations are used by researchers in various disciplines. These models are in general binomial models—binomial in the sense that they focus their attention on two causal variables: 1) that part of the population who have already adopted the innovation, and 2) the rest of the population who are potential adopters. However, these models have a serious limitation in that the potential adopter population is assumed to remain constant over time. This paper presents some modified binomial innovation diffusion models that incorporate dynamic potential adopter populations. Moreover, the developed models are applied to some case studies, and their superiority in forecasting the time pattern of diffusion is also included in this presentation.     

Forecasting technology diffusion: a new generalisation of the logistic model

The present research proposes a new generalisation of the logistic model aiming at technology diffusion forecasting. Regarding criticisms and failures reported in the literature to apply logistic function for long-term forecasting, in our work we focused on short-term accuracy of forecast. To formulate the model, based on mathematical approximation, at first the differential equation governing the diffusion process is found and then by solving the derived differential equation, the forecast function is obtained. In all steps, mathematical tools from numerical analysis are used. We compared the New Generalized Logistic Model with eight of the most renowned models in the literature. The model led to more accurate fits and forecasts than those obtained from other models we applied for comparison.     

Two generalized rational models for forecasting innovation diffusion

This article demonstrates the development and use of two Generalized Rational Models I and II (GRM I and II) representing innovation diffusion. Specifically, the GRM II covers the same area as the NSRL model, which includes the Coleman and the Blackman/Fisher-Pry models, while the GRM I covers the same area as a modified NSRL model (mod. NSRL), also introduced hereby, and including Floyd, Blackman/Fisher-Pry, Sharif-Kabir and Exponential models. Both the GRM I and the GRM II provide a form of differential equation which always has for a solution a fact which cannot be met when dealing with the NSRL and mod. NSRL models.Some applications are presented, first to illustrate the wide applicability and the usefulness of the models and second to demonstrate the alternate use of the GRM I and mod. NSRL, and GRM II and NSRL models, which usually approximate very well.     

Forecasting technology diffusion with the Richards model

The Richards model has a shape parameter m that allows it to fit any sigmoidal curve. This article demonstrates the ability of a modified Richards model to fit a variety of technology diffusion curvilinear data that would otherwise be fit by Bass, Gompertz, Logistic, and other models. The performance of the Richards model in forecasting was examined by analyzing fragments of data computed from the model itself, where the fragments simulated either an entire diffusion curve but with sparse data points, or only the initial trajectory of a diffusion curve but with dense data points. It was determined that accurate parameter estimates could be obtained when the data was sparse but traced out the curve at least up to the third inflection point (concave down), and when the data was dense and traced out the curve up to the first inflection point (concave up). Rogers' Innovation I, II and III are discussed in the context of the Richards model. Since m is scale independent, the model allows for a typology of diffusion curves and may provide an alternative to Rogers' typology.     

A study on the relationship between technology diffusion and new product diffusion

This study investigates the relationship between “technology diffusion” and “new product diffusion”. We define “technology diffusion” as a knowledge spillover process, which is represented by patent citation, and “new product diffusion” as the spread of a new product that has been developed by the application of patented technology. To investigate the relationship between the two types of diffusion, we use patent citation data of code division multiple access (CDMA) technology and market sales data of mobile phones in South Korea for the analysis. The results show that the diffusion of technology through patent citation could be successfully explained by empirical analysis, for which the Bass diffusion model was used. Moreover, we can find out if technology diffusion can be the leading indicator of a new product's diffusion before its launching; in other words, before the commercialization of the patent.     

A product diffusion model incorporating repeat purchases

A logistic-based model for forecasting the rate of product diffusion given aggregate time series data was constructed. The model differs from earlier models based on fitting the logistic to aggregate data in that it includes a submodel to separate replacement demand from first-time sales. We fit the theoretical model to data and show that forecasts will be significantly more accurate using this model instead of the logistic curve.     

A stage-based diffusion of IT innovation and the BSC performance impact: A moderator of technology–organization–environment

Information system (IS) innovation is an important resource to link to firm performance according to the resource-based view. However, the diffusion of IS innovation is often the crux of the final successful use and thus, critical to realizing firm performance. The diffusion process is dynamic and complex in nature. The innovation diffusion theory (IDT), which is inherent with a multi-stage analysis, can provide insight to its implementation. Little research has examined IT-enabled firm performance for a multi-stage analysis. Moreover, empirical studies have shown inconclusive results in assessing IT value. The balanced scorecard (BSC) is a hybrid performance measure system with four indicators. Grounding on the IDT and BSC, this study proposed a novel research model to examine the relationships between a stage-based diffusion structure and the four BSC indicators. Further, the technology–organization–environment (TOE) was defined as a moderator between them as it is critical in determining the adoption of technology innovation. The results indicated that IT value is realized differentially in different forms of performance indicators across different diffusion stages. Time-lag effect is also important for a well-realized financial performance.     

Adaptive diffusion models for the growth of robotics in New York state industry

Forecasts are developed for the diffusion of robotics in the state of New York through the year 2015. The chief objective is to compare static approaches with dynamic models for forecasting diffusion processes of various time horizons. Results for a Bass-Mansfield model are compared to those for a dynamic time-varying parameter model. The results indicate the advantages and disadvantages of a robust heuristic approach which smooths data as opposed to providing an optimal fit.     

A multi-attribute diffusion model for forecasting the adoption of investment alternatives for consumers

This article develops a diffusion model that incorporates potential adopters' perceptions of the relevant innovation attributes in explaining the rate of adoption of an innovation. Data from 14 investment alternatives currently available to consumers are used to develop a multi-attribute diffusion model for forecasting the acceptance of a potential investment alternative. Limitations and further extensions of the proposed model are also discussed.     

Modelling the faddish, fashionable and efficient diffusion of agricultural technologies: A case study of the diffusion of wool testing technology in Australia

This study investigates the nature of innovation diffusion in an agricultural context. The dominant agricultural diffusion models assume that an economically rational choice is made to adopt or reject agricultural technologies. However, recent studies of agricultural innovation highlight the ‘irrational’ and potentially ‘inefficient’ nature of the diffusion in this context. To investigate how and why agricultural technologies are adopted or rejected, we examine the diffusion of wool testing technologies in the Australian wool industry using the Bass diffusion model and Abrahamson's diffusion and rejection typology. The results show that diffusion of agricultural innovation is not simply an efficient choice made to close observable performance gaps. The findings suggest that the adoption of inefficient innovations and the rejection of efficient innovations can be driven by an adopter's social context, powerful external influences and imitation within an adopter group and that these drivers change over time, suggesting an evolutionary social process underlies the diffusion of agricultural technologies.     

Polynomial innovation diffusion models

In an earlier paper [42] the authors presented a comprehensive evaluation and extensions of available causal models of “binomial type” for describing the time pattern of the innovation diffusion processes. The binomial models are based on the assumption that the entire population can be divided into two groups—adopters of an innovation and the potential adopters—such that eventually everyone adopts the innovation and an innovation once adopted is never rejected. However, many examples can be cited where this assumption is unrealistic. Therefore this paper presents some polynomial innovation diffusion models that are less restrictive compared with the binomial models. The paper also shows the link between the polynomial diffusion process and the multilevel technological substitution process.     

Modelling a dynamic market potential: A class of automata networks for diffusion of innovations

Innovation diffusion processes are generally described at aggregate level with models like the Bass Model (BM) and the Generalized Bass Model (GBM). However, the recognized importance of communication channels between agents has recently suggested the use of agent-based models, like Cellular Automata. We argue that an adoption or purchase process is nested in a communication network that evolves dynamically and indirectly generates a latent non-constant market potential affecting the adoption phase.Using Cellular Automata we propose a two-stage model of an innovation diffusion process. First we describe a communication network, an Automata Network, necessary for the “awareness” of an innovation. Then, we model a nested process depicting the proper purchase dynamics. Through a mean field approximation we propose a continuous representation of the discrete time equations derived by our nested two-stage model. This constitutes a special non-autonomous Riccati equation, not yet described in well-known international catalogues. The main results refer to the closed form solution that includes a general dynamic market potential and to the corresponding statistical analysis for identification and inference. We discuss an application to the diffusion of a new pharmaceutical drug.     

Innovation forecasting

Technological forecasting is premised on a certain orderliness of the innovation process. Myriad studies of technological substitution, diffusion, and transfer processes have yielded conceptual models of what matters for successful innovation, but most technological forecasts key on limited empirical measures quite divorced from those innovation process models. We glean a number of concepts from various innovation models, then present an array of bibliometric measures that offer the promise of operationalizing these concepts. Judicious combination of such bibliometrics with other forms of evidence offers an enriched form of technological forecasting we call “innovation forecasting.” This provides a good means to combine technological trends, mapping of technological interdependencies, and competitive intelligence to produce a viable forecast. We illustrate by assessing prospects for ceramic engine technologies.     

Innovation diffusion in a heterogeneous population

One of the assumptions made in modelling innovation diffusion is that of complete mixing of prior and potential adopters. In the present paper we give a deterministic model for diffusion of innovation in a population for which this assumption is inappropriate. The population of interest is divided into two spatially separated groups. Diffusion in the first group near the single centre of innovation is assumed to follow the logistic model. A modified model is proposed for the second group. Individuals in this group are assumed to receive information from previous adopters in both the groups. The model is fitted to data on diffusion of crossbred goats in villages around Narayangaon, a town about 70 kilometers from Pune in Western India.