The concept of innovation

Innovation is considered essential for the growth of both production and productivity (^{OECD, 2005}) and is mainly reflected in the profits or business value through increased sales or cost reduction (^{COTEC, 2007}). In the food industry, growth usually discontinuous, so some companies adopt a logic of innovation to maintain or improve their market position (^{Fortuin et al., 2007}). The agrifood innovation in the sector is then a prospective strategy to overtake competitors, generating individuals, organizations, team managers and conditions that favor or reduce that constrain it. Innovation as a strategy and the food industry as the focus of its application, form the subject called agrifood innovation.

The concept of innovation refers to a new process or product, or improvers result of the incorporation of new and existing knowledge. It is an evolutionary and nonlinear process that requires communication and intensive collaboration between different actors, companies and organizations (^{Tödtling and Trippl, 2005}). The use of innovations in a sustained manner, leading to improve competitiveness and to promote the development of territories (^{Méndez, 2002}; ^{Ayalew et al., 2015}).

The innovation, and application of knowledge, aspires to the social and economic improvement, primarily through the development of enterprises (^{COTEC, 2001}; ^{Alburquerque-Llorens, 2008}). This development must be oriented to inclusive (^{Foster and Heeks 2013}; ^{Heeks et al., 2013}) that allows upgrades and the value generated reach the population traditionally excluded from the benefits of innovation.

The individuals and companies have discussed the process of innovation, from conception of ideas, until these ideas are transformed into goods and services that generate value. Thus arises the conception of innovation as any changes brought about by it based on knowledge and generating value (^{COTEC, 2007}).Another orientation in the definition of innovation (^{OECD, 2005}) is its conceptualization as the introduction of a new or significantly improved product (good or service), a process, a new marketing method or a new organizational method, in the internal practices of the company, workplace organization or external relations. Under a neo-Schumpeterian approach, the innovation process is a process of deliberate and specific learning by companies (Burgueño and Pittaluga, 1994). In these definitions stand the elements of novelty, diversity and value generated, to which can be added inclusiveness to be submitted innovation processes.

In the beginning (^{Schumpeter, 1934}) he defined innovation as a process of creative destruction where the profit motive drives constant innovation, which involves not only the introduction of new products, but the establishment of new rules, and marketing successful new materials and components, the introduction of new processes, opening new markets and the introduction of new forms of marketing. This course takes the approach of innovation as something new for the company (^{OECD, 2005}) that can be developed internally, acquired, or combined process.

The systemic approach to innovation (^{Klerkx et al., 2009}) recognizes that agricultural innovation is not just about adopting new technologies, but also requires a balance between new technical practices and alternative forms of organization, for example, markets, holding land and distribution of benefits. This recognizes the multi- disciplinary and trans-discipline as conditions that favor innovation, accepting that the integration of knowledge is a condition that favors innovation. The Frascati Manual provides a wide range of research and development activities that could be considered as innovations developed in various fields such as medicine, social sciences or militia (^{OECD, 2002}). Therefore, it can be studied innovation in management or aimed at improving society through value creation in business economics of technological change (^{López, 1998}).

The knowledge is a means and generating value to the innovation process. By not generate value, through many changes generated and applied knowledge, one could speak of initiatives or developments, but not innovation. Innovation involves a culture, where change is recognized as an intrinsic part of daily work, knowledge is valued as the basis for these changes, and the value generated as the common goal of the company.

The changes are a common part of the innovation process so that companies develop areas of innovation management as part of its core business areas (^{Fortuin et al., 2007}). Management processes focus innovation (^{Cooper, 1999}) to the realization of the right projects, correctly. Innovation is a constant change (^{OECD, 2005}) which must be understood as part of the management of innovation.

Once understood the concept of innovation at company level, various studies take up the approach to innovation at the network level and chain (^{Omta, 2002}) which can be interpreted as the need for institutional innovations (^{Perez, 2010}) or organization or innovation management (^{Frambach and Schillewaert 2002}). The chain vision and network can recognize innovation as a social process where interaction is part of the innovation process itself (^{OECD, 2005}; ^{Perez, 2010}; ^{OECD, 1997}). The interaction has been studied under an approach of "integrator" that relationships conducive to innovative production process (^{Klerkx et al., 2009}; ^{Howells, 2006}; ^{Dhanarai, 2006}).

Some research (^{Howells, 2006}; ^{Fortuin et al., 2007}; ^{COTEC, 2007})) indicate that the integration of a team focused on innovation and good communication are key to generating business value factor. These teams are integrated into the diffusion of innovation is inside the enterprise, outside, or both. Diffusion is part of the innovation process (^{Rogers, 1995}; ^{Valente, 1996}; ^{Wejnert, 2002}; ^{OECD, 2005}; ^{Klerkx et al., 2009}) by which elements for discussion of various approaches to the diffusion of innovations are discussed; from those designed linearly (^{López 1998}) to those designed under a systemic approach (^{Klerkx et al., 2009}) or so-called innovation chain or interactive (Landau and Rosenberg, 1986).

Innovation under the look of system

Understanding innovation under a systemic approach favors the design of public policies more comprehensive to promote the effective strengthening innovative capabilities through close cooperation of innovation actors (^{Chung, 2002}; ^{Salom and Carrasco, 2003}; ^{Lamprinopoulou et al., 2014}). From this approach, the analysis of a system must consider its structure, processes, inputs (input) and output (output).

Innovation systems (SI) are defined as the set of actors, organizations and institutions that interact in the generation, dissemination and use of economically useful-knowledge- and new production processes (^{Fischer, 2001}).

In order to understand these processes of innovation initially worked at national level (nation-state) coining the term national innovation system (SNI) (^{Lamprinopoulou et al., 2014}) conducting various studies (^{Dutrénit et al., 2010}; ^{UNESCO, 2010}; ^{Klerkx et al., 2012}; ^{Ayalew et al., 2015}; ^{Läpple et al., 2015}). ^{Chung (2002)} suggests understand and analyze the SNI as a set of subsystems classified by sectors or regions so that they can establish differentiated innovation policies consider specific features of the territory.

Innovation and territory

From the geographical perspective, studies of proximity between the actors and the effects that this agglomeration promotes regarding the exchange of knowledge, have a clear relationship with innovation processes (^{Méndez, 2003}).

The territory is a geographical area that is previously formed by a system of objects and a set of actions that work in a unified manner. There is a clear fusion o factions and objects. The geographical space can be considered as something that participates in the condition of the social and physical, a mixture, a hybrid. "Objects and actions are gathered in a logic that is at the same time, the logic of past history (date, its material reality, its original cause) and the logic of today (its operation and its significance present). The geographical significance and value of geographic objects (technological artifacts) are from the role in the social process" (^{Santos, 2000}).

The territory fulfills the functions of regulation, habitat, production and information on the ecological network (^{De Grood et al., 2002}), and that is where innovations emerge. In addition, the territorial perspective provide greater effectiveness to innovation policies to be located in its implementation and execution space (^{Alderete, 2013}). This vision of the territory should consider the dimensions related to the flow of information, communication and technical development to characterize it properly and to use it as the basic unit of public policy management (^{Martinez, 2010}).

Regional innovation systems

The relationship between the territory and innovation under the gaze of system seeks to describe this subtle, and in some cases, tacit exchange of knowledge applied to new products framed in a region (^{Asheim and Coenen, 2005}; ^{Cooke and Gómez, 1998}; ^{Li, 2009}; ^{Mendez, 2002}; ^{Salom and Carrasco, 2003}; ^{Uyarra and Flanagan, 2009}).

The approach of regional innovation systems (SRI) believes that the regions are meaningful unit of analysis to study innovation to be organized into administrative units that must directly support innovation (^{Uyarra and Flanagan, 2009}).

However, the definition of region as the unit of analysis is variable because the dimensions that encompass the concept of region (sociocultural, functional and administrative) may be present or not uniformly in the territory. Therefore, it is necessary to address the problem with tools that consider the fundamental role of the territory in the incubation of innovative activities aimed at local development (^{Rózga, 2003}).

Understanding the dynamics of a production system as a whole in a given locality is to recognize the role played by regional actors to then become directly involved in the production, so that leverage the competitive advantages of the location that originate in the bonds of cooperation and improvement of existing knowledge (^{Alderete, 2013}).

The specific innovation

^{Diaz-Jose (2013)} proposes specific innovation systems (SEI) as a planning strategy of innovation that goes from 'bottom-up' to generate and analyze information properly on players SEI and relationships that occur among them. Therefore, they are defined as the coordinated heterogeneous actors interacting for the generation, dissemination and application of new economically useful knowledge in the development of an activity in a territory or region as a whole.

Applying the concept of SEI to analyze the smallholder agriculture or agriculture self-supply can relate a specific crop, the actors and their relationships, the characteristics of the territory so that it can study the level of innovation they employ and the circuits that follow or not the processes of knowledge management.

Methodological problems in innovation

The identification of actors in a territory is as variable as the territory itself. There are complex territories in their geographical characteristics as well as social, economic, cultural, and political, among others. To this end they have created, designed, adapted and used different methodologies. However, using a single methodology is not sure that is the right or to allow compliance with the ultimate goal of the research, so a conglomeration of methodologies, which together, to identify the actors in the field is necessary and know the context in which the productive systems in the territory they operate.

To identify local actors is proposed to start with the methodology snow ball (^{Bergek et al., 2008}; ^{Klerkx and Leeuwis, 2008}; ^{Kilelu et al., 2011}).

This methodology involves the prior identification of a small group of actors to identify others with the same characteristic desired. The new members will reference identified turn to other actors and so on to achieve the final identification of actors depending on the purpose of the investigation (^{Muñoz et al., 2004}).

^{Hanneman (2005)} points out a weakness of this methodology the fact that actors who are isolated will not be referenced. Nor does it guarantee the identification of all actors connected within the study population. These reasons show the need to complement this approach with other tools to cover these weaknesses.

Social mapping as a tool to build knowledge collectively, can complement the methodology snowball making an approach to the actors that takes into account geographical, social, economic, cultural and historical space (^{Martínez, 2010}). This methodology allows, from an interdisciplinary perspective, linking actors with the perception of its territory (^{Diez et al., 2012}; ^{Vélez et al., 2012}).