Examining the effects of collaboration in R&D/innovation projects via game theory analysis

Although it has been defined in the 20th century as we use it today, innovation has actually always been at the center of human activity as one of the engines of economic growth. The importance of innovation increases and the time between cycles decreases day by day since Industrial Revolution. One of the important components of innovation appears to be R&D, which is widely supported by states aiming not to fall behind in competition. The role of collaborations in ensuring that these works are carried out in a more effective, efficient and sustainable way become more visible in last decades. The aim of this thesis is to model the establisment of cooperation in order to obtain higher social benefit from R&D studies, how the parties will behave according to the cooperative game theory, and to explore the dynamics in this regard. The actors in this game is defined as university, industry and public R&D institutions. Their decisions on whether or not to take part in a highly uncertain, complex and multi-disciplinary R&D study has been examined as a coalition game. The game theoretical model is established by extending a model in the literature, in which new product development cooperation between two identical companies was examined. Extension is in the number and characteristics of players, problem type from Nash Bargaining to coalitional game, from NPD to pre-competitive R&D. In order to examine the effects that may help in predicting the success of the R&D project, which is one of the model parameters, logit and mixed-effect logistic regression models are inspected via automatic model selection procedure using the data of 170 completed R&D projects. In the selected model, the extension of the R&D projects' duration positively affected the perception of success. Also, the effects of collaborations on the success of R&D projects vary depending on the type of these projects. Therefore, the type, founding type and aim of R&D projects should not be ignored in determining the institutions that will cooperate. The perception of success and cooperation dynamics were examined in an exploratory study with 16 experts (three of whom were within the scope of the pilot study and excluded). The game-theoretic model aims to ensure that all potential outputs can be obtained from the project by ensuring the fair distribution of the rewards, given to the different R&D outputs to increase the social benefits. In the game theoretic analysis, the appropriate sharing ratio in a model in which the aforementioned three players try to increase the total profit by addressing the issues of trust, coordination, learning together and absorbtive capacity has been examined with the concepts of coalition solutions. The parties involved in the project contracts, especially the funders, can increase the grand coalition formation probability and thus the total gains by distributing the potential revenues of the different coalition structures according to the sharing ratios calculated by the Shapley value. Our model considers both the monetary value and the effects of other cooperation dimensions such as trust, knowledge investment, compatibility of competencies. These dimensions show that the gains to be achieved via learning will be more than the extra costs due to cooperation. Therefore, the players may act in favor of cooperation and increase the potential outputs of the project, thereby increasing the social benefit. The outline of the thesis is as follows: In the first section, the basic concepts related to R&D, the characteristics and difficulties that distinguish these studies from others, and the direct and indirect outputs of R&D activities are mentioned. Basic concepts are mentioned in order to explain the relevant terminology and to introduce the types of projects to be used in the analyses. The characteristics and difficulties are mentioned to introduce concepts such as project success, probability of success, absorptive capacity, complementarity and knowledge spillovers, which will be used in the supportive analyzes and in the model. R&D outputs are described because of their role in the pay-off calculation in our mathematical model. In the second section, information is given about the importance, advantages and disadvantages of collaborations in terms of R&D/innovation studies. The main motivation of our study is to examine the dynamics of university-industry-public R&D institution cooperation in R&D projects with different purposes or types. Therefore, the types of R&D cooperations, the motives that lead to or avoid from cooperation, the factors that will affect the success of cooperation and potential risks in a collaboration are addressed. Then, the characteristics of the cooperation partners, which are the subject of the thesis, their motivation for cooperation and the stages of cooperation are mentioned. In the third section, general information about the definition and development of game theory is given. Its collaborative branch and coalition games, which are used as analysis methods in the thesis, are explained. Some other game-theoretic studies in the field of R&D and innovation are also briefly mentioned. In the fourth section; the purpose, importance, limitations and assumptions of the research were mentioned. Then, the methods and findings of the logit and mixed-effect logistic regression analyses and the exploratory study, which are carried out to obtain information about the mathematical model and some of its assumptions, are mentioned. In the final section, main results are discussed. The findings obtained by logit and mixed-effect logistic regression analyzes contribute to the efforts to understand the differences between "project" and "R&D project". For example, in the project management literature, exceeding project deadlines is a feature that is often mentioned when describing unsuccessful projects, but on the contrary, it can be perceived as an indicator of success in an R&D project. Intangible elements such as raising human resources, creating knowledge and a culture of cooperation are important for an R&D project. In addition to the concrete outputs determined by the project goals, facilitating new projects is also an important component for R&D success; an R&D project is often perceived as successful to the extent that it contributes to future projects. Deviation from the budget or timeline is not an indicator of failure for the R&D project, it is almost by default. Collaborations in successful R&D projects are beneficial when carried out with the minimum number of partners required and where the share to be received/contributions to be given can be clearly determined, and in cases when it is inevitable for the competencies to complement each other. In the presence of unclear contributions to the project, extra partners, large differences in the level of commitment or goals of the parties, or conflicting competencies, the complexity of R&D collaborations increases and various managerial problems may arise. Our game theoretic model aims to ensure that all the potential outputs can be obtained from R&D and increase social benefits by ensuring the fair distribution of the rewards via Shapley value.