Organic molecules (i.e. carbon-based) have opened a new and rapidly-growing industrial field in the optoelectronic market bringing to this field a new dimension of thinness and flexibility. In this context, this thesis has focused on one particular building block of the vast and emerging field of organic electronics: the organic thin-film transistor (OTFT) which uses organic compounds as semiconductor. Whereas the OTFT-based circuits are not meant to compete with the silicon-based high-end industry (micro-processors...), their performance have already reached levels enabling their use in potential applications such as displays (e-paper, LCD, OLED) or radiofrequency identification (RFID) tags. The continuously growing number of available organic molecules exhibiting conductive, semi-conductive or insulating properties combined with the number of available deposition/patterning methods (e.g. gravure printing) gives more flexibility to the technology. These additional degrees of freedom raise two main questions: How to identify the most suitable OTFT platform for a given application and how to estimate its potential, as for instance in, of digital circuits? This thesis targets to answer to those questions. For this purpose, several OTFT platforms have been screened and their performance have been discussed and compared through standard figures of merit. The self-aligned nano-imprinted technology has demonstrated state-of-the-art sub-micrometer OTFTs on 4-inch flexible substrates. This made this platform the most suitable candidate for developing the potential evaluation framework. For that purpose, a static model suitable for the sub-micrometer OTFTs has been developed which embeds almost all known electrical aspects of OTFTs. Then the device-to-device discrepancy often observed in OTFTs has been studied and statistical modeling methods introduced. This allowed the simulation of sub-micrometer inverters performed with commercially available tools. Next, a statistical method has been developed to evaluate the potential of the sub-micrometer OTFTs for digital applications. Whereas the method concludes that these sub-micrometer OTFTs are not mature enough to make complex digital circuits, this methodology is technology-independent and may thus serve as a basis to characterize unipolar-logic printed electronics and be further extended to complementary-logic circuits. Last but not least, an automation effort has been undergone all along this thesis in order to increase the throughput for such demanding data analysis. The main outcome of this task is a user-friendly multi-analysis and parameter extraction platform.