Building public innovation for industrial risk prevention and crisis management: Genesis and development of a unique collaborative innovation

Collaborative innovation: An important driver to overcome barriers to public innovation

Until the 1980s, the role of the public sector consisted mainly in granting the best possible conditions for the private sector to create innovation (Sørensen, 2012). The situation changed radically from the end of the 1990s, due to multiple factors that have driven public innovation. One of the most important factors is the growing scarcity of resources in the public sector, which leads public actors to perform with less, and thus look for innovative solutions. Another relevant factor concerns so-called ‘wicked problems’ (Rittel & Webber, 1973), i.e. problems that are particularly complex and very difficult to solve, such as climate change, poverty reduction, and public security. They require innovative solutions because they cannot be solved by public administrations using existing solutions. A third factor is the pressure exerted upon public administrations by globalization, which urges competitive dynamics, encouraging governments and regions that are able to innovate and penalizing less virtuous administrations (Sørensen & Torfing, 2012b). Finally, the increasing demand by the general public for tailor-made services – for example, in the health sector – pushes public administrations to find novel solutions (Bommert, 2010).

Public innovation can be understood as innovation in the public administration or as innovation by the public administration. The former entails organizational and strategic changes in the public administration, especially in pursuit of the innovation imperative to reform and reinvent public services. ² The latter entails innovation as the result of public administration’s action through changes that can be intentional, accidental and episodic (Eggers & Kumar Singh, 2009) or unintentional, sometimes also referred to as ‘bricolage’ (Fuglsang, 2010: 67). In both cases, public innovation can be prevented by several barriers, such as the absence of measures to encourage innovation (Landeau, 1993), the proliferation of rules and procedures in state bureaucracies that produce constraints for innovation (Moore, 2009), and the often risk-averse culture of public administrations (Mulgan, 2007). These barriers can lead governments to ‘one-off’ changes through a ‘big-bang approach’ (Eggers & Kumar Singh, 2009: 6), potentially preventing administrations from adopting larger strategies and becoming serial innovators (Albury, 2005).

In order to overcome public innovation barriers there are at least three different drivers. None of these is intrinsically superior to another. As a matter of fact, there is no best way to produce innovation and the question is rather to adopt the best strategy according to the scope of innovation that is at stake (Hartley et al., 2013). The first driver is based on the neo-Weberian state (Pollitt & Bouckaert, 2011), in which innovation is mainly produced within public organizations thanks to the behavior of skillful leaders, public managers, and employees. The second driver, which refers to the principles of New Public Management, assimilates the public sector to the private one, thus encouraging innovation as the result of competition among actors (Sørensen, 2012).

The third driver is collaborative innovation. This approach implies forms of collaboration that can successfully spur innovation because the innovative process is opened up, integrating a wide range of actors with their assets in the innovation cycle, thus increasing the quality and quantity of innovations (Bommert, 2010). Collaborative innovation can be considered as a new research area (Torfing, 2013). Nonetheless, several scholars have provided empirical evidence of the crucial impact of collaboration on public innovation in many different countries at a national level all over the world, including the United States (Agranoff, 2014) and European countries such as Denmark (Carstensen & Bason, 2012) and Norway (Corwin et al., 2012), at a regional level (Dossou-Yovo & Tremblay, 2012; Montin et al., 2014), and at a local level (Agger & Sørensen, 2014). Moreover, many case studies show that collaborative innovation is not limited to a restricted number of domains but spans from health-care systems (Pedersen & Johansen, 2012) to water management (Bressers, 2014), and a wide range of other sectors (Agolla & Van Lill, 2013).

Collaborative innovation is no panacea (Hartley et al., 2013). As a matter of fact, collaboration among actors is sometimes unsuccessful due to coordination problems (Swan & Scarbrough, 2005). Moreover, collaboration can be prevented by power asymmetries, mistrust, and opportunistic behavior (Ansell & Gash, 2008). Nonetheless, collaborative innovation can be considered as an effective way to identify and define problems so as to develop new solutions and facilitate ‘trust-based circulation and cross-fertilization of new and creative ideas’ (Sørensen & Torfing, 2012b: 5). Moreover, collaborative innovation can successfully overcome barriers to public innovation by implementing new solutions thanks to resource exchange and coordination (Sørensen & Torfing, 2011).

SIMAGE: A multidimensional innovation for industrial risk prevention and crisis management

Industrial risks are particular risks for the population that are due to human activities in the industrial sector, such as a fire or explosion involving chemical substances, ³ and are the subject of specific European risk policies. ⁴ They can constitute a catastrophic event for human health, as was the case with those that occurred in Seveso (1976), Bhopal (1984), and Toulouse (2001), to mention but a few examples. Many devices have been created to prevent such accidents. They can be divided into three main categories: 1) systems that monitor the quantity of chemical substances in the air; 2) devices that analyze available data; 3) systems for crisis management (Da Ronch et al., 2010).

SIMAGE is one of these devices, aimed at protecting the health of the population and workers. The system has two main functions. The first is to prevent industrial risk. The second is to manage the crisis in the event of an accident. It was designed in 2001, tested in 2006, and has been fully operational since 2007. ⁵ It was installed in the major Italian industrial site of Porto Marghera (Zucconi, 2009), which spans approximately 2,000 hectares and is divided into three different areas: a refinery, chemical plants, and shipyards. This site is well known at a national level for three main reasons: employment issues and trade union mobilizations (Chinello, 1996); chronic environmental pollution due to chemical substances released in the air, water, and land (Bortolozzo, 1998); a trial concerning the death of workers due to the exposure to monomer vinyl chloride (Bettin & Dianese, 2003). In comparison, issues relating to industrial risk have become visible rather recently, especially since an accident that occurred in the Dow Chemical plant of Porto Marghera in November 2002 in which there were no victims but extensive material damage.

SIMAGE can be considered as a multidimensional innovation as defined by Hartley (2005). First, it is a service innovation, performing an unprecedented service for the population that lives in the surroundings of the industrial site, integrating three heterogeneous tasks in a single system, while existing devices usually perform one task at a time, or two at most ⁶ (Da Ronch et al., 2010). As a matter of fact, SIMAGE monitors the quantity of pollutants in the air, analyzes the relevant data gathered in the monitoring phase in real time, and manages emergency procedures in case of an accident. Second, SIMAGE is a product innovation. Indeed, this new service for the population was made possible by a set of technological instruments (Crivellari, 2015), which radically modified existing ones. Moreover, SIMAGE is an innovative public good that has no possible restriction of access or exclusion mechanisms and is, at the same time, absolutely non-rivalrous since it can be used by an indefinite number of individuals without reducing its availability. Therefore, it can be considered as a ‘pure public good’ (Leach, 2004), as opposed to non-excludable and non-rivalrous ‘impure public goods’ ⁷ that can become congested or where exclusion dynamics may be possible.

The region of Veneto ⁸ decided to finance the SIMAGE project in 2001 for a total amount of four million euros. The governor of the region, while acknowledging the established importance of the chemical industries, stated publicly that the region had opted for the construction of ‘a new Porto Marghera’ characterized by logistics, transportation, and services. ⁹ SIMAGE was to be considered as a tool implemented by local policies that would reduce industrial risk locally, thus contributing to foster a transition of Porto Marghera from heavy chemicals to the tertiary sector (Crivellari, 2016). The task of designing SIMAGE was assigned to ARPAV. This environmental agency, created in 1996, performs environmental prevention through its technical structures throughout the region and has administrative and technical autonomy. Along with twenty other regional agencies in Italy, ARPAV is part of the system of environmental prevention and performs both technical and policy-making tasks. ¹⁰

ARPAV involved the plants of the site in the project. The plants were (and still are) gathered in an association named EZI (Ente Zona Industriale), which was formed in 1924 and has, among its missions, those of protecting the interests of the plants and providing site management of technical services. The public agency acted as ‘convener’ (Sørensen & Torfing, 2012b: 7) to initiate the process, identify relevant stakeholders, and bring them together. ARPAV’s decision to interact with the site’s plants was not a foregone conclusion, because up to that point their relationship had been one between the controller (ARPAV) and the controlled (chemical plants). As a matter of fact, the public agency performs inspections of the plants and at least one technician of the environmental agency is present every day of the year in the perimeter of the site in order to perform these tasks. Moreover, ARPAV has judicial police powers to enforce Italian risk prevention legislation.

Method

The empirical data for this article have been gathered through a sociological research conducted with qualitative techniques, namely semi-structured interviews and analysis of paper and electronic documents. I conducted 49 interviews in three phases. In the first exploratory phase in 2009, I conducted 11 interviews with informants drawn from three groups. The first group comprised local government technicians in industrial risk departments of the municipality and the province of Venice. The second group included activists from a group of citizens mobilizing against chemical risk in Porto Marghera. The third group comprised journalists from different local newspapers who had covered environmental and risk-related issues. In the second phase in 2012 and 2013, I carried out 33 interviews. 15 of these were conducted with people directly engaged in the design, development, and functioning of SIMAGE, and with actors responsible for industrial risk prevention and crisis management, as they are specified in the Italian law on industrial risk n° 334 of 1999. 18 interviews were conducted with citizens of two local groups linked to industrial risk. The first was called ‘The Permanent Assembly Against Chemical Risks’ (APCPC) and was formed in 2002 by citizens mobilizing against industrial risk. The second group was called ‘Group for Information, Protection and Safety of the Italian Civil Protection’ (GIPS) and was formed by the municipality of Venice in 2003. In the third phase in November 2018, I conducted 5 interviews in order to follow up recent local developments in local chemical risk regulation, in the APCPC’s mobilization, and GIPS’ activities.

A digital recorder was used for each of the interviews after obtaining the explicit consent of the informants, except for one interview because the interviewee denied consent (in this case my notes represent the interactions). I then used the Audacity system for mastering and editing digital audio data (audacityteam.org), and fully transcribed the interviews in Word format. The interviews lasted ninety minutes on average. Due to the diversity of interviewees and the complexity of the issues, I opted for interviews guided by a list of points according to the specific role of the informant and did not create standardized questionnaires. After interviewing a respondent, I asked her/him for referrals for other potential respondents. The quotations cited throughout the text have been translated from Italian from the verbatim transcription of the interviews. 14 of the interviews have been conducted and transcribed by Riccardo Emilio Chesta, whom I thank for the collaboration.

The empirical material has also been collected in paper and electronic format. I collected materials provided by interviewees, such as flyers, newspaper clippings, brochures, intra-agency communiqués, and corporate documents. I also used a newspaper review provided by the ARPAV’s press relations officer for a total of 75 newspaper clippings on SIMAGE and industrial risk in Porto Marghera, published between July 4, 2007 and November 27, 2011 in the local newspapers La Nuova Venezia, Il Corriere del Veneto di Venezia e Mestre, and Il Gazzettino di Venezia. I also used the newsletter of the Permanent Assembly Against Chemical Risk published at www.margheraonline.it, which has been reconstituted in a 783-page document in pdf format.

The first phase: From idea to technological innovation through cooperation

The first step in the SIMAGE project was to monitor the concentration of chemical pollutants in the air within the perimeter of the industrial site. The task to build a particular device to accomplish this function was assigned to a group of ARPAV technicians led by a manager specialized in industrial chemistry. The innovative idea in this phase was to design a system for monitoring air pollutants to prevent an explosion or fire. This idea was coherent with Italian law n° 334 of 1999, which defines industrial risk as an event such as an emission, fire, or explosion that occurs during the activity of a plant, causing immediate danger to human health.

The idea was innovative because it was completely different from any underlying devices already in existence in Porto Marghera, which performed the monitoring of pollutants in the air to check atmospheric pollution. As a matter of fact, the scope was focused on acute events and was therefore clearly different from the traditional monitoring of environmental pollution due to the chronic release of substances in the air, water, and land. Moreover, it was centered on protection of the health of the population from imminent danger, and not on environmental issues. Therefore, the substances to be monitored were different from the ones checked for environmental purposes and were chosen according to chemical properties such as explosiveness and flammability.

An instrument to monitor air quality pollutants already existed and was available in the market. It was called DOAS (Differential Optical Absorption Spectroscopy). The DOAS device principally comprised a beam emitter and a beam receiver, which recorded the quantity of a given substance in the air at a given moment (Platt, 2006). This type of device was designed to monitor urban pollution (Edner et al., 1993) and was not designed for industrial risk prevention. Therefore, emitters and receivers were always positioned just a few meters from one another. The ARPAV innovation consisted in the proposition to modify the specifics of the DOAS system, so that emitters and receivers would be located a few hundred meters away from one another:

DOAS did not exist as an instrument in the territory. As a matter of fact, the firm producing DOAS [. . .] would reason in terms of much shorter distances than those required at Porto Marghera. [. . .] So, DOAS, which was designed to check air quality, becomes, with SIMAGE, ‘on or off’ on the presence of a substance within a certain perimeter. If I have a substance, it means that an emergency is in progress. And that’s the way an instrument for air quality has become an instrument of crisis management. (Interviewee 25)

In the fall of 2003, the region of Veneto issued a call for bids. It was awarded to Sartec, the distributor in Italy of the system produced by the Swedish firm Opsis, based on the strength of its innovative proposal to use non-traditional specifics for measuring pollutants in the air.

In a similar way, ARPAV proposed to modify the specifics of other technological tools used for monitoring chemical substances: gas chromatographs. These devices, usually employed to analyze the concentration of substances in gaseous samples, were modified to be used in risk prevention at the site of Porto Marghera. Hence, the public agency modified already existing tools, thanks to the skills of its technicians and their knowledge of the field. As the ARPAV manager responsible for this phase explains:

Everybody makes gas chromatographs, but not the gas chromatographs that measure the light organic compounds in a single operation. And here, our technical know-how counted for a lot. Not even the firm that sold it [the gas chromatograph] knew that they could do this! We invited them to our laboratory [. . .] with the result that they then sold it with the new specifics that they had learned from us. (Interviewee 24)

As many scholars point out, a new idea is not to be confused with an innovation. Innovation is novelty in action (Altshuler & Zegans, 1997) and differs from invention because it is not only a new idea (Bessant, 2003), but also a new practice. The contribution provided by EZI was crucial for the transition from idea to innovation. In fact, ARPAV did not possess all the resources to build the system alone. The agency needed, in particular, to install the devices in a private area, within the perimeter of the chemical firms. This implied authorizations and permits and the building of support infrastructure, such as plinths, for example, to install emitters and receivers. But, until then, the relationship between controller and controlled had been strictly formal and based on mutual mistrust. EZI played a key-role as a mediator to facilitate cooperation between the parties. The director of the association told us:

We were committed to giving ARPAV support and the link to private companies. Authorizations to occupy the land, free loans, etc., were necessary. Then, also, the whole construction that leads to the building of the device. (Interviewee 25)

As a result, this cooperation allowed ARPAV and the private companies to interact on a voluntary basis in the prevention of industrial risk in a way that went beyond a formal controller-controlled relationship. The result was a form of innovation that went beyond barriers to innovation caused by mistrust and consisted in the production of innovation through cooperation between different actors of the public and private sectors who interacted and communicated with one another, albeit without sharing the risks or costs. In this case, the risks and the cost of the innovation lie entirely with the public actor.

The third phase: Crisis management through synergy

The third and last stage of the SIMAGE project was crisis management. Contrary to the first and second phases, which were intended to prevent a chemical accident, this phase was designed to manage a crisis after an industrial accident had occurred. Many of the institutional actors in the territory already had specific tasks defined by industrial risk regulation, namely the Prefect, the municipality of Venice, the region and the province of Veneto, the fire brigade, and the Civil Protection. In case of an accident, the local fire brigade intervenes within the site to control the explosion or extinguish the fire and secure the area. Outside the perimeter, the External Emergency Plan (EEP) coordinated by the Prefect includes measures such as the evacuation of the population with the intervention of the Civil Protection and the local authorities.

ARPAV technicians decided to equip a room ad hoc, so that these actors could converge in one place in case of an accident, thus accelerating emergency procedures and optimizing interinstitutional communication. The room, which is called the ‘operational room’, is still located in the main building of the fire brigade in Mestre, a few kilometers away from the industrial site. Should an accident occur, the actors are supposed to meet there and manage the crisis, relying on the technical work of ARPAV’S and EZI’s technicians, who can give all the information available about the accident, the characteristics of the plants involved, and data on the environmental follow-up.

SIMAGE also performs the diffusion of the communication to the population. Italian legislation already enforces information procedures for communicating to citizens about chemical accidents. Therefore, numerous loudspeakers are disseminated in the municipality territory. SIMAGE introduced specific technological devices, such as large screens that are located along highways and roads with heavy traffic. Moreover, some information terminals were installed in areas highly frequented by the public, such as Venice’s airport. These devices were devised to communicate updates on the incident, measures to be undertaken by the population and, eventually, the announcement of the end of the emergency. The RIALTO system was also created. This system conveys information on industrial accidents through television and radio, interrupting ongoing programs with specific messages to the population.

The third phase of SIMAGE was made possible by synergy among the various actors with responsibilities related to industrial risk. ARPAV and EZI expanded their mutual collaboration to other actors who had a technical role thanks to their competencies in chemistry, chemical engineering, and knowledge of local plants. The various actors acknowledged the role of these technicians and trusted them. ARPAV and EZI, albeit preserving their autonomy, became a unique interface for the other actors, thus blurring the frontiers between the private and the public sector. As one of the employees of the municipality of Venice told us:

In my opinion, what is quite clear is that there is a synergy of all those who are interested in the problem [of industrial risk], that has grown more and more. That is, I know those who work at ARPAV, the Fire Brigade, etc., very well . . . and they know me. There is a lot of communication. [. . .] This is a work that has lasted for years. It has not been always this way. (Interviewee 12)

Innovation through collaboration between the public and the private sector

SIMAGE is a public innovation based on collaboration for the creation of a new service for the population that prompted an evolution from a formal controller-controlled relationship to collaboration. ARPAV’s managers have acknowledged this evolution:

I would say that the most interesting thing is that, number one, there have not been big accidents and, number two, the wall has fallen a bit . . . I can tell you that if something occurs to a plant, it is not simple to tell it, to share it with the inspection authority, to be completely transparent. Here, in comparison to twenty years ago, things have radically changed. There is no suspicion towards the inspection authority. In the end, a dialogue exists, there is a shared management. Really, the wall of reluctance that existed until ten years ago has fallen. (Interviewee 16)

As another ARPAV’s manager sums up: ’We have now reached the point where we say: “What is the problem? Let’s define it together”’. (Interviewee 26). The private companies also acknowledged it:

SIMAGE has first of all allowed a new era to open up for public administration, precisely because there is no longer a command-and-control relationship; instead there is a relationship of, let us say, mutual collaboration, remaining of course within the division of each role. It has allowed companies to understand the problems that public agencies have to deal with in managing this issue [industrial risk]. It has allowed the public agencies to understand the problems that companies have in managing the plants. Each puts itself in the other’s shoes. (Interviewee 25)

Interaction between actors evolved through cooperation and negotiation, gradually becoming a collaboration that consisted in sharing tangible and intangible resources. In 2005, ARPAV and EZI signed an agreement to jointly finance SIMAGE maintenance expenses, which amounted to approximately 700,000 euros annually. They also shared information, which is considered to be very important for the process of developing innovation (Bland et al., 2010; Sørensen, 2012) and managed to effectively orchestrate knowledge to produce a complex innovation (Dougherty & Dunne, 2011). Moreover, it is a collaboration based on a common scientific and technical language, which the ARPAV and EZI technicians created during the formal meetings and informal exchanges. This language was based on technical expertise in calculus, probabilities, and technicalities concerning industrial risk, understood as the probability that chemical substances can cause an accident with negative repercussions for workers within the perimeter of the site and for the population surrounding it.

This innovation is characterized by the inclusion of several actors. A wide number of institutional actors that are responsible for particular tasks were included, as often happens in crisis management given the high complexity of the problem that requires the effort of networks of actors (Li & Goodchild, 2010). This innovation is also characterized by the exclusion of citizens. SIMAGE performs only a unilateral communication, in which citizens are the passive receiver of a message (Da Ronch et al., 2010). Therefore, citizens are the recipients of the service provided by SIMAGE, and they pay for it through taxation, but are not involved in the design and building of the system. It is a choice that is assumed by the public actor and grounded on a precise communication strategy that prefers to keep a very low profile. As explained by an ARPAV manager:

The real issue is not to illustrate SIMAGE, but to understand if you, the citizen, think that I [the ARPAV] am a third party in governing the territory. In other words: ‘Do you trust my analysis, or do you think that another analysis that is different from mine is necessary?’ This is the most delicate aspect. This led us to work in silence: Time will tell who’s right. [. . .] Because of the fear of ‘hurting ourselves’, because we feel vulnerable from a communication point of view [. . .] it’s better to remain silent. (Interviewee 26)

It should be noted, though, that citizens have not formally raised demands for inclusion in the design and building of SIMAGE. The members of the spontaneous citizens’ association against chemical risk formed in November 2002 – APCPC – have mainly focused on the request for a referendum on risk prevention, namely on the delocalization of phosgene. Therefore, as some of the respondents from the APCPC told us (Interviewees 22, 23), criticism of SIMAGE has been limited to the delays in the building of the system in comparison with the original project. The members of the Group for Information, Protection and Safety of Civil Protection are volunteers and do not protest against industrial risk and the way it is managed by the public authorities, being engaged instead in procedures of information provision to the population about crisis management (Interviewees 13, 37, 42).