M a J H U a A R R A A K I E P B 1 ( ( t r d a i 2 s c t f 4 a T h 0 Process Safety and Environmental Protection 117 (2018) 372–382 Contents lists available at ScienceDirect Process Safety and Environmental Protection journa l h om epage: www.elsev ier .com/ locate /psep aking the links among environmental protection, process safety, nd industry 4.0 ose Alcides Gobbo Junior ∗, Christianne M. Busso, Simone Cristina O. Gobbo, enrique Carreão NESP – São Paulo State University, Production Engineering, 17033-360, Bauru, SP, Brazil r t i c l e i n f o rticle history: eceived 16 December 2017 eceived in revised form 14 May 2018 ccepted 16 May 2018 vailable online 21 May 2018 eywords: ndustry 4.0 nvironmental protection rocess safety a b s t r a c t Industry 4.0 is an emerging concept in production systems and is described as a concept that encompasses technologies such as the Internet of Things, big data, cyber-physical systems, and smart objects. Industry 4.0 will present new challenges and opportunities for process safety and environmental protection (PSEP) researchers and managers. There is a gap in the literature in identifying the main concepts related to industry 4.0 and PSEP, as well as the potential integration between these subjects. In line with this gap, this paper focuses on identifying and systemizing information regarding the integration of PSEP and industry 4.0 concepts and technologies. While identifying relationships, we also address non-existent intersections between keyword co-occurrence networks in PSEP and industry 4.0. The main outputs of this research are: (a) identification and discussion of potential connections between PSEP with industry 4.0 concepts; and ibliometric networks (b) an in-depth discussion of potential benefits in integrating industry 4.0 concepts and technologies into PSEP fields. It can be concluded that there is much more collaborative research between environmental protection and industry 4.0 than between process safety and industry 4.0. Also, environmental protection research is characterized by a wide variety of research themes and multidisciplinary endeavours, in 0. nstitu contrast with industry 4. © 2018 I . Introduction The themes of process safety and environmental protection PSEP) have implications for various areas of production systems Gysin, 1976). Process safety is the common global language used o communicate the strategies of hazard identification and analysis, isk assessment and evaluation, safety measures, and safe critical ecision making (Khan et al., 2015). Another driver to address haz- rdous installations and major accidents has been the regulations ntroduced by governments for environmental protection (Hutton, 017). Companies’ decisions directed towards sustainable production ystems will require a consistent and continuous procedure, espe- ially in the PSEP fields. Industry 4.0 has been considered one of he major trends in production systems, with clear implications or sustainability in organizations (Jabbour et al., 2018a). Industry .0 is an emerging trend in production systems and is described as concept that encompasses technologies such as the Internet of hings (IoT) (Lee and Lee, 2015), big data (Lee et al., 2014; Hauhang ∗ Corresponding author. E-mail address: gobbo@feb.unesp.br (J.A. Gobbo Junior). ttps://doi.org/10.1016/j.psep.2018.05.017 957-5820/© 2018 Institution of Chemical Engineers. Published by Elsevier B.V. All rights tion of Chemical Engineers. Published by Elsevier B.V. All rights reserved. et al., 2015), cyber-physical systems (CPS) (Lee et al., 2015), smart manufacturing (Kang et al., 2012), and smart objects (Fortino et al., 2013; Gobbo Junior et al. 2017). Thus, this paper argues that integration among concepts and technologies of industry 4.0 and PSEP would unlock the poten- tial for sustainable production systems. An additional benefit to do research in these subjects would come from the already exist- ing interaction between emerging topics in industry 4.0 and PSEP functions. A resulting cross-fertilization of ideas may facilitate the identification of novel approaches to risk reduction and environ- mental protection (Hutton, 2017). Therefore, the research in the PSEP arena will need to be integrated in the emerging cutting-edge discussion of industry 4.0. Despite the fact that process safety and environmental protec- tion researchers have been pioneers in the study of the impact of automation on safety, industry 4.0 will present new challenges and opportunities for process safety and environmental protection researchers. Angell and Klassen (1999) proposed an integration of environmental issues into research in productive systems. Early research integrating industry 4.0 and sustainable production con- cepts includes the works of Jabbour et al. (2018a, 2018b) and Jabbour et al. (2017). reserved. https://doi.org/10.1016/j.psep.2018.05.017 http://www.sciencedirect.com/science/journal/09575820 http://www.elsevier.com/locate/psep http://crossmark.crossref.org/dialog/?doi=10.1016/j.psep.2018.05.017&domain=pdf mailto:gobbo@feb.unesp.br https://doi.org/10.1016/j.psep.2018.05.017 Envir t i s i i a s i b f ( w a e a d 4 e e i a s i 5 2 2 p t c t T a a a i m c c s n o o t f p T m ( ( p ( o ( i d J.A. Gobbo Junior et al. / Process Safety and Although there are very few studies on the frontier of the rela- ionship between these subjects and systematic reviews on the solated themes, no studies have attempted to integrate, through a ystematic analysis, concepts in PSEP and industry 4.0. Another gap n the literature is the lack of research with a systematic approach n effectively identifying the main concepts related to industry 4.0 nd PSEP, as well as the existing intersections among these subjects. In line with this gap, this paper focuses on identifying and ystemizing information regarding the integration of PSEP and ndustry 4.0 concepts. While pointing out early intersection points etween these subjects, we will also address possible directions or future integrative research into these fields, following Jabbour 2013). Consequently the primary question guiding this research is: hat are the possible gaps and intersections points between PSEP nd industry 4.0 concepts? Taking this into account, the outline of this paper is to identify merging research fronts in PSEP and industry 4.0. A secondary im is to identify possible gaps and intersections as well as new irections in integrating the research themes in PSEP and industry .0. This identification of possible gaps and intersections will be laborated by means of one very specific but influential type of vidence: topological measures in co-occurrences of keywords. The paper’s sections are organised as follows: the main theoret- cal background on PSEP and industry 4.0 is described in Section 2; pproaches for visualization of bibliometric networks and exten- ions are described in Section 3; the research method is presented n Section 3; the results are discussed in Section 4; and in Section some conclusions are presented about this work. . Theoretical background .1. Process safety: concepts and principles Safety is a relative concept that must be understood in the resence of some hazard or risk. The concept of risk is related o both hazards created by humans and those created by nature; onsequently, safety constitutes an ability to reduce or eliminate he likelihood of hazardous events occurring (Antonsen, 2009). he main focus of risk analysis is to support decision making by ssessing and quantifying the risks associated with the operation nd design of a technical system (Aven, 2016). Risk assessment nd risk management provide important contributions in support- ng the decision-making process. Therefore, risk assessment and anagement are a set of principles and methods developed to con- eptualize, assess, and manage risk and recognized threats (also alled hazards) (Aven, 2016). A growing interest in the process safety field is the concept of afety culture. The existing research in the field supposes a con- ection between safety culture and safety. The conceptualization f safety culture is by no means conclusive. Aligned with this view, rganizational culture may help to create safety by being a medium o disseminate safety-critical knowledge (Antonsen, 2009). There- ore, the inherent safe concept could be obtained in a process or roduct design that avoids hazards instead of controlling them. his stage would be achieved through four main methods: mini- izing (reducing the amount of hazardous material), substituting replacing one material with another of less hazard), moderating using less hazardous process conditions), and simplifying (design rocess to be less complicated and therefore less prone do failure) Heikkilä, 1999). Process safety is identified as an integral part of process devel- pment and manufacturing rather than an “add-on” to the process Gibson, 1999). Process safety differs from occupational safety as t solely focuses on preventing and mitigating major process acci- ents such as fires, explosions, and toxic releases. Process safety onmental Protection 117 (2018) 372–382 373 assessment/management includes several essential steps. Though every step is equally important, hazard identification, risk assess- ment, and management can be considered as the key steps of process safety management (Bahr, 1997). The primary objective of process safety is to prevent the unwanted release of highly hazardous chemicals/biological agents into locations, which could expose humans to serious hazards. Process safety management is an approach to evaluate processes that have the potential to cause catastrophic incidents such as fires, explosions, or toxic releases (OSHA, 2000). Extensive industrial automation and computer control create many new issues in process safety, which have been extensively discussed in the literature (Hendershot, 2006). This is mainly due to human factors. Examples of different applications of automation to avoid human errors can be found in Hendershot (2006), Cameron et al. (2017), and Taylor (2017), not only to avoid safety hazards but also to automate Hazop’s analysis or prevent maintenance. 2.2. Environmental protection: concepts and principles Environmental protection is the practice of protecting the natural environment of the individual, organizational, and gov- ernmental levels, for the benefit of both the environment and humans (Kraft, 2017). Traditionally, environmental protection has been considered a public interest and external to the private sphere (Mazurkiewicz, 2004). The idea behind this concept is that the bio- physical environment has been degraded and governments have begun to place restraints on activities that cause environmental degradation (White, 2017). Since its popularization, the concept of environmental protection has become closely associated with the concept of sustainable development. However, gradually the focus is shifting from sustainable development to the multiplicity of sustainability and analyses of the paths in which they are shaped (Haughton and Counsell, 2004). Discussion concerning environmental protection often focuses on the role of government, legislation, and law enforcement. How- ever, in its broadest sense, environmental protection may be seen to be the responsibility of all stakeholders, including industry, environmental, and community groups (Kraft, 2017). Gradually, environmental decision-making processes are evolving to reflect this broad base of stakeholders and are becoming more col- laborative in many countries. The stringency of environmental regulations affects eco-innovation as firms respond to stricter environmental regulations with higher levels of eco-innovations (Kesidou and Demirel, 2012). Peachey (2008) says that environmental protection manage- ment involves a complete assessment of the full impacts of power source choices and a conscious decision about what environmen- tal impacts and mitigation measures are going to be made. The terms environmental quality, air quality, soil quality, and water quality invariably connote the status or condition of each condition that relates to requirements of wildlife or to human requirements and/or preferences (Johnson et al., 1997). These are related to the role of government in environment pro- tection. Urban air pollution poses a significant threat to human health, property, and the environment throughout both the devel- oped and developing parts of the world. The issue of urban air quality is receiving increasing attention as a growing share of the world’s population is now living in urban centres and demanding a cleaner urban environment (Gurjar et al., 2008). One impact of air emissions is that they also contain significantly high levels of sulphur components—heavy metals and fly ash being generally of greater regional concern because of direct health and environmen- tal impacts (Peachey, 2008). All of them need to be monitored to ensure that they are in control under the specifications of govern- 3 Envir m c h ( r t t f 2 m a a c i H e s a w T m t s r t e t u o t i ( c r s m t c p o t o a t e d o m F a p d m n topics, which contemplate title, abstract and keywords. This search has been done at SCOPUS database, as also and there are no doc- uments for this combination of keywords. Therefore, we suppose that an individual selected keyword search highlights the keywords Table 1 Data collected from Web of Science. Web of Science Keywords used in this Industry 4.0 Environmental Process Safety 74 J.A. Gobbo Junior et al. / Process Safety and ent institutions. There are different technology devices to this ontrol. Analysing the subject of environmental protection automation as been used to control water quality, air quality, and heavy metals Zhang et al., 2016; Mishra et al., 2011), among other monitoring equired to meet the specifications of government. One applica- ion of data automation in the environmental protection field is he data mining approach. Nawahda (2016) applied data mining or predicting surface ozone levels with traffic information. .3. Industry 4.0: concepts and technologies Industry 4.0 is a promising approach based on integration of anufacturing and business processes in the company’s own oper- tion and supply chain. Technical aspects of these requirements are ddressed by the application of the generic concepts of CPS, the IoT, loud computing, big data, smart objects, and manufacturing to the ndustrial production systems (Lee and Lee, 2015; Lee et al., 2014; auhang et al., 2015; Lee et al., 2015; Kang et al., 2012; Fortino t al., 2013; Gobbo Junior et al. 2017). CPS are defined as the joint work of computers and physical ystems, which allow the integration between physical processes nd objects in order to connect machines and devices in a net- ork of production lines (Lee and Lee, 2015; Lee et al., 2015). echnically, CPS consist of four main elements: the cybernetic ele- ents (computers and software) that enable physical elements o become smart objects; the computational networks that allow mart objects to communicate with each other and with the envi- onment; sensors and actuators embedded in smart objects; and he environmental capture and transmission information (Gunes t al., 2014). The IoT refers to the interconnectivity of physical objects and he Internet through wireless or cable networks and the use of nique identification codes for the purpose of mutual exchange f data (Atzori et al. 2010, Gubbi et al. 2013). As a consequence, he IoT depends on a number of technologies (Lee and Lee, 2015): dentification technologies such as radio-frequency identification RFID), which allow the identification of objects and automatic data apture; wireless network sensors to monitor physical and envi- onmental conditions as well as provide information about objects uch as their location and movements; middleware that allows ediation between software and other applications, moving or ransforming information between different platforms, communi- ation protocols, and operating systems; cloud computing for the rocessing and storage of a large amount of data generated in the rganization; and application software that enables robust data ransmission and interaction between objects and people. These smart production systems are implemented with the bjective of creating a horizontal integration of digital networks nd a vertical integration of these networks with the manufac- uring systems (Kagermann et al. 2011, Prause, 2015, Strandhagen t al., 2016). Smart manufacturing systems objectives include the evelopment of CPS and dynamic production networks with the bjective of building up flexible and open value chains in the anufacturing of mass customization products (Ramsauer, 2013). ive characteristics are associated with the term, namely context wareness, modularity, heterogeneity, interoperability, and com- ositionality (Mittal et al., 2017). The smart factory concept also escribes smart production systems that operate in the environ- ent of a factory, while they are able to connect with a global etwork of similar production systems (Mittal et al., 2017). onmental Protection 117 (2018) 372–382 3. Research method Bibliometric networks are composed of nodes—publications, journals, researchers, or keywords—and edges, which indicate rela- tions between pairs of nodes (Van Eck and Waltman, 2014). Once referred to as ‘science mapping’, bibliometric networks have been admitted to the mainstream of bibliometric research (Shibata et al., 2008). Many different approaches have been proposed for visu- alizing bibliometric networks. The most commonly studied types of bibliometric network relations are citation, co-citation, bib- liographic coupling, keyword co-occurrence, and co-authorship. Popular visualization techniques include distance-based, graph- based, and timeline-based approaches (Van Eck and Waltman, 2014). Our focus in this research is on the co-occurrence network approach. Networks of co-occurrences of keywords have also been studied extensively (Luiz et al., 2016). Keywords of a publication can be extracted from the author-supplied keyword list, or from even the title or abstract (Van Eck and Waltman, 2014). In some cases, especially in the older literature, keywords are restricted to individual words, but in other cases they also include terms con- sisting of multiple words. For examples of early work on keyword co-occurrence networks, we refer to Callon et al. (1983) and Peters and Van Raan (1993). Google Scholar database provides a broader coverage in compar- ison with Scopus and ISI Web of Science (Harzing and Alakangas, 2016). Scopus database therefore is more extensive than ISI Web of Science, however in the later, Natural Sciences and Engineering (NSE) journals are well overrepresented (Mongeon and Paul-Hus, 2016). NSE is an important discipline for PSEP and industry 4.0 researchers and have the biggest share of journals in ISI Web of Science in comparison with Scopus. The source of the keyword data used for this research was the ISI Web of Science database, which differs from others by its wide content and by grouping an updated database, which includes arti- cles in the preview phase and very recent publications. The data were collected in November 2017 by using a keyword search in the database search engine. The keywords selected were “industry 4.0”, “environmental protection” and “process safety”. The selec- tion fields considered were Title, Abstract, and Keywords, with no restrictions of period, language, journals, or type of documents. To explore the link between the themes, bibliographic data were downloaded for all publications and are listed in Table 1. The ISI Web of Science database supports various file formats. We used the text (txt) format to capture details of all documents, and the bib- liographic data were downloaded in batches of 500 publications at each time (Table 1). Therefore, we downloaded three batches for “industry 4.0”, 58 batches for “environmental protection”, and four batches for “process safety”. A previous search, linking all the three selected keywords, was conducted in VOSviewer and for the combination “Environmental Protection” and “Process Safety” and “Industry 4.0” there are no records. The analysis has been done by research Protection Documents identified with the keywords 789 28879 1873 Number of Keywords 62 500 124 J.A. Gobbo Junior et al. / Process Safety and Environmental Protection 117 (2018) 372–382 375 menta c t i i e T a b s e w e t fi e u s t t W a T v b w i 4 s r y a e 1 g e w s c m u r a Fig. 1. Distribution of publications of Environ o-occurrence network for each topic. The missing literature for his combination of selected keywords helps to justify the need to dentify possible gaps and intersections points between PSEP and ndustry 4.0 concepts. After the collection of data, the database in text extension was ntered in VOSviewer to analyse the co-occurrence of keywords. he search for keywords co-occurrence was referenced to the title nd abstract, with a minimum of five co-occurrences and with inary counting, disregarding repetitions. VOSviewer has some pecial text-mining features and also allows editing, correcting rrors, and adding missing bibliographic data elements. Therefore, e also applied a filter available in VOSviewer to not consider irrel- vant terms such as articles and prepositions. After the selection by he software, the least relevant items were excluded and manual ltering was used to exclude words considered not relevant (for xample: findings, conclusions, etc.). As a limitation, this software treats direct citations relations as ndirected relations; however, mapped networks may consist of everal thousand nodes. To construct co-occurrence networks of erms, VOSviewer uses text-mining functionality to extract tex- ual data from titles and abstracts of publications (Van Eck and altman, 2014). The VOSviewer filter selects all word sequences nd plural noun phrases and converts them into singular ones. herefore, the software calculates for each noun phrase a rele- ance score (Van Eck and Waltman, 2014). To identify similarities etween keywords in PSEP and industry 4.0, spreadsheet software as used. A list of keywords was built considering each period and ts respective clusters. . Results and discussion The initial analysis is about the distribution of publications pre- ented in Fig. 1. This figure indicates an incipient number of articles elated to the topic in the first years, growing significantly in recent ears. There is a more interest in the early 1990s for this theme, prob- bly due to the world meetings and climate conferences on the nvironment, like the conferences in Toronto in 1988, Geneva in 990, and Rio de Janeiro in 1992, which placed the theme on lobal agendas. Environmental protection is a theme that has been xplored since 1970; however, industry 4.0 is a recent theme, for hich a first publication was dated 2012, and process safety has hown a strong presence in academic research since 1991. The graphical analysis of VOSviewer depicted the researchers’ ountry origin network related to each keyword. Figs. 2–4 show the ain countries and the links between the researchers in different niversities. In the visualization presented in Figs. 2–4, each circle epresents a country. Large circles represent countries that have large number of publications. Small circles represent countries l Protection, Process Safety and Industry 4.0. with fewer publications. In general, the closer that two countries are located to each other in the visualization, the more strongly they are related to each other based on bibliographic coupling. In other words, countries that are located close to each other tend to cite the same publications, while countries that are located far away from each other usually do not cite the same publications. Colours indicate clusters of countries that are relatively strongly related to each other. Analysing Fig. 2, it is possible to visualize that in the environ- mental protection field the countries that have larger influence were the U.S. and China. Already in the case of research about process safety the country that develops more research about this subject is the U.S. (Fig. 3), and it is possible to say that England, China, Germany, and Canada are emerging forces and have strong influence along with the U.S. Finally, research about industry 4.0 started in Germany, which together with China remains the main country that has researched this subject (Fig. 4). 4.1. Co-occurrence network of terms (keywords) In the visualization presented in Figs. 5–7, keywords networks were grouped into clusters. In the cluster of environmental protec- tion (Fig. 5), 24 keywords clusters were identified, with the largest one related to water and air quality and heavy metals. The second large cluster of keywords is related to environmental management and the third with hazards identification and assessment. It is a dense network comprising more than 500 keywords. In the cluster of process safety (Fig. 6), 12 keyword clusters were identified, with the largest one related to fault tree analysis; the following clusters were auditing, decision making, chemical pro- cess, and inherent safety design. This network has a smaller density in comparison with environmental protection and comprises 124 keywords. The higher density of the environmental protection key- words network could be explained by the larger number of papers published yearly (589 papers per year) since the 1970s, in com- parison with process safety (38 papers per year). Because of the much newer history of industry 4.0, with the first paper published in 2012, it has a more sparsely connected keyword network and comprises 62 keywords. PSEP also has a smaller number of clusters (10); the first cluster is related to CPS and automation, the second to manufacturing technologies, and the third to digitalization issues. Tables 2–4 show the result of keyword co-occurrence searches for the 10 main words (by number of occurrences) and the total link strength for each keyword. It is possible to observe that between the most co-occurrence keywords in PSEP and industry 4.0 there are no similarities. The analysis of similarities shows that there are more inter- sections between environmental protection and industry 4.0 (19 keywords co-occurrences) than with process safety (6 keywords 376 J.A. Gobbo Junior et al. / Process Safety and Environmental Protection 117 (2018) 372–382 Fig. 2. Researcher country network – Environmental Protection. try ne c c c t P g Fig. 3. Researcher coun o-occurrences). Possible gaps were identified in Table 5 and are omposed of the non-existent intersection between keywords o-occurrences in industry 4.0, process safety and environmen- al protection. There are also similarities between the following SEP and industry 4.0 keywords in Table 5: automation; inte- ration, maintenance, optimization and simulation. Most of the twork – Process Safety. papers using this combination of keywords in industry 4.0 and PSEP were published in English, and country authors were mainly from China, Germany and USA. Documents for industry 4.0 and process safety were classified mainly in the engineering, computer science and automation control systems fields. Environmental protection has a more diverse documentś classification (proceeding papers, J.A. Gobbo Junior et al. / Process Safety and Environmental Protection 117 (2018) 372–382 377 Fig. 4. Researcher country network – Industry 4.0. or the a i s w i t Fig. 5. Keywords and cluster by color f rticles, reviews, book chapter, etc.) and fields (engineering, chem- stry, materials science, environmental science ecology, computer cience, water resources and energy fuels). Probably this result as due to the higher number of articles and keywords observed n environmental protection in comparison with the other clus- ers (industry 4.0 and process safety). Therefore, environmental search of “Environmental Protection”. protection reveals more diverse application of the similar key- words (automation, integration, maintenance, optimization and simulation) and greater possibilities of application of industry 4.0 concepts, due to this multidisciplinary characteristic. Linking concepts between these arenas would provide new directions for research. One example, one would be to explore the 378 J.A. Gobbo Junior et al. / Process Safety and Environmental Protection 117 (2018) 372–382 Fig. 6. Keywords and cluster by color for the search of “Process Safety”. color u m p a i b Fig. 7. Keywords and cluster by se of information technology, especially cloud computing and data ining, in the process safety field. Cloud computing would sup- ort organizations in storing large amounts of data, related to risk ssessment, analysis, and management activities and applications n the cloud. Organizations could also improve their process safety y using data mining techniques and genetic algorithms in the for the search of “Industry 4.0”. large amount of safety-related data stored in the cloud for pattern discovery and solution optimization. Although we have not found in our research a relationship between the process safety cluster and the keyword data mining, Qi et al. (2012) and Cameron et al. (2017) showed that this tool can be a way to organize data, find pat- terns, and propose changes to anomalies by developing an active J.A. Gobbo Junior et al. / Process Safety and Envir Table 2 10 main Keywords of search “Environmental Protection”. Keyword Occurences Total Link Strength environmental protection 1509 2225 sustainable development 521 899 risk assessment 445 1104 environment 442 1021 china 296 560 air pollution 269 605 sustainability 230 470 water quality 227 543 heavy metals 209 435 mercury 160 388 Total Clusters 24 Table 3 10 main Keywords of search “Process Safety”. Keyword Occurrences Total Link Strength process safety 318 358 process safety management 83 137 risk assessment 46 75 safety 43 65 process hazard analysis 27 46 safety culture 26 51 risk analysis 26 44 inherent safety 26 37 psm 25 63 risk management 22 31 Total Clusters 12 Table 4 10 main Keywords of search “Industry 4.0”. Keyword Occurrences Total Link Strength industry 4.0 359 388 internet of things 50 99 cyber-physical systems 49 92 big data 31 61 smart factory 28 56 smart manufacturing 18 39 cloud computing 18 34 iot 15 33 cps 14 31 industrie 4.0 11 17 Total Clusters 10 Table 5 The similarities of words among the three clusters. Similarities of Keywords Cluster Industry 4.0 Cluster Environmental Protection Cluster Process Safety automation x x x cloud computing x x not exist data mining x x not exist genetic algorithm x x not exist human factors x not exist x information technology x x not exist innovation x x not exist integration x x x internet of things x x not exist logistics x x not exist maintenance x x x manufacturing x x not exist optimization x x x rfid x x not exist security x x not exist simulation x x x smart grid x x not exist supply chain x x not exist supply chain management x x not exist sustainability x x not exist onmental Protection 117 (2018) 372–382 379 knowledge-based retrieval system that provides users necessary information about chosen chemicals by integrating data mining with text mining to record and analyse the incidents. This data stor- age can also be done by cloud computing and can be optimized by genetic algorithms. The human factor concept is concerned with the alignment between the user, equipment, and the environment. Despite exten- sive studies within process safety for the role that human factors have on safety, there is a current gap in research in relation to environmental protection. Adoption of innovations in the PSEP field, such as the IoT and technologies related to security (safeguarding of personal data) should be further explored. Fabiano (2017) affirmed that future research in innovation is required to cope with new chal- lenges and connected safety. The use of sensors connected to the Internet to monitor physical and environmental conditions would allow manufacturing organizations to proactively and effec- tively reduce equipment and environment-related hazards. Logistic routes would be optimized according to safety and environmental indicators with RFID support. The smart grid concept involves the use of an improved information system to match energy consump- tion with production (Gobbo Junior et al., 2017) and could reduce the risk associated with equipment failure or malfunctioning. These initiatives of integration between industry 4.0 and PSEP concepts could be expanded from the organization perspective to the supply chain, therefore leading to progress towards sustainability. By constructing the matrix in Table 6 we relate and propose potential connections between the main keywords (by number of co-occurrences) of PSEP with industry 4.0. An explanation of the relationships of these keywords follows Table 6. The number of keywords used in Tables 2–4 was reduced; excluded keywords included acronyms (e.g., IoT) or words in different languages with the same meaning (e.g., industrie 4.0). Process safety management (PSM) would benefit from the IoT and CPS. These technologies could help to avoid accidents and prevent future loss through the collection of data from processes and objects, such as manufacturing equipment, being possible to identify potential sources of hazard failures. Big data and cloud computing technologies would be beneficial to PSM by applying a large amount of data to plan, monitor, and control variables related to human factor errors and hazards of industry equipment. The set of principles and methods of risk assessment, analysis and management (process safety), and risk assessment (environ- mental protection) would benefit from sensors and RFID tags in smart objects informing operators of risks related to materials or components in manufacturing environments. Risk assessment could be more assertive with the technology of data mining, cloud computing, and the IoT, because it is possible to use historical data to give more confidence to predictions related to probability and potential loss. The IoT and CPS would provide similar benefits to risk assessment and analysis as the ones described in PSM. The process hazard analysis approach could be improved through the use of both cloud computing and big data, because these technologies would allow the collection, storage, sharing, and fast processing of large amounts of data. Cameron et al. (2017) affirmed that to improve the effectiveness of hazard identifica- tion and scenario definition, a systems approach and a computer supported/semi-automated methodology is paramount. Sensors embedded in objects could obtain hazard data and transmit infor- mation through RFID. This information would be accessed in real time, to improve and accelerate the decision-making process. The concepts of safety, safety culture, and inherent safety could especially benefit by two main characteristics associated with smart manufacturing systems: awareness and interoperability. Increasing safety awareness in manufacturing operations could be a beneficial consequence of this concept adoption. Interoperability 380 J.A. Gobbo Junior et al. / Process Safety and Environmental Protection 117 (2018) 372–382 Table 6 Matrix of relationships between the main Keywords of Process Safety, Environmental Protection and Industry 4.0. Process Safety Industry 4.0 Process Safety Management AND Big Data; Cloud Computing; Internet of Things; Cyber Physical Systems Risk Assessment; Risk Analysis; Risk Management Big Data; Cloud Computing; Internet of Things; Cyber Physical Systems Process Hazard Analysis Big Data; Cloud Computing; Internet of Things; Cyber Physical Systems Safety Culture; Safety; Inherent Safety Smart Factory; Smart Manufacturing Environmental Protection Industry 4.0 Sustainable Development; Sustainability AND Smart Factory; Smart Manufacturing Risk Assessment Big Data; Cloud Computing; Internet of Things; Cyber Physical Systems Environment Big Data; Cloud Computing; Internet of Things; Cyber Physical Systems t w p i b S a a t p e m t t a b t i t l c c m q ( o t w m C f i p c c t i ( e o i t 5 i e Air Pollution; Water Quality; Heavy Metals; Mercury China ould reduce failures in systems operations interfaces, reducing otentially related hazards, accidents, or material losses and mak- ng the production systems more inherently safe. The sustainable development and sustainability concepts could e improved by the smart factory and manufacturing systems. mart production systems in the environment of a factory or network create horizontal and vertical integration of digital nd manufacturing systems, building up flexible and open sys- ems. These systems are more oriented to services, instead of hysical products, therefore relying less on material usage. Less nvironmental impact would be a result, as also would be a ore distributed manufacturing system, therefore reducing Co2 ransport-related emissions. The environment could be controlled from the use of IoT echnologies, from sensors to apps, in order to plan, monitor, nd control environmental factors related to land management etween harvesting rotations (Jabbour et al., 2018b), the applica- ion of IoT-related technologies to develop smart cities concepts n urban environments (Kehua et al., 2011; Su et al., 2011), unin- errupted monitoring of environmental changes and precision ocation of pollution sources (Lu et al., 2015). The use of big data and cloud computing would optimize pro- esses and improve information gathering to predict and better ontrol water quality, air pollution, and contamination by heavy etals or mercury. Automation has been used to control water uality, air quality, and heavy metals (Zhang et al., 2016). Nawahda 2016), for example, applied data mining for predicting surface zone levels with traffic information. However, sensors and RFID echnology embedded in smart objects connected to the Internet ould efficiently collect information and allow control of environ- ental conditions in real time. Artificial intelligence embedded in PS would allow automating the decision-making process, there- ore leading to rapid decision making in the case of potential threats n production systems. Increasing attention has been paid to urban and agricultural soil ollution with heavy metals in China (Dong et al., 2018). Mining is onsidered to be one of the most significant sources of heavy metal ontamination in China (Li et al., 2014). China’s pesticide indus- ry was ranked the second largest in the world. The adoption of ndustry 4.0 concepts and technologies in China, such as big data Lu et al., 2015) or smart manufacturing, could lead to improved nvironmental conditions in that country, specifically the quality f soil, among other problems. Thus, we further suggest propositions for future studies that ntegrate the concepts and technologies of industry 4.0 and PSEP opics. . Final remarks The results from this research indicate that there are more ntersections between the keywords co-occurrences clusters of nvironmental protection and industry 4.0 than process safety. The Big Data; Cloud Computing; Internet of Things; Cyber Physical Systems Big Data; Cloud Computing; Internet of Things; Cyber Physical Systems; Smar Factory; Smart Manufacturing intersections between environmental protection and industry 4.0 concepts and technologies are in the fields of supply chain manage- ment, technology, production, and human safety and sustainability. New avenues for research could emerge from the combination of non-existing links between the main keywords in the clusters of environmental protection, process safety, and industry 4.0. It can be concluded that there is more collaborative research in environ- mental protection and industry 4.0 than with process safety. Also, environmental protection research is characterized by a wide vari- ety of research themes and multidisciplinary work, in contrast with industry 4.0. Geographical inequality in the publication output is identified as a point of concern, because a geographical concentra- tion was observed, mainly in process safety and industry 4.0. 5.1. Contributions of the work This work contributes to integration between PSEP and industry 4.0 concepts and technologies. The main outputs of this research are (a) the matrix of the relationships between the main key- words in PSEP and industry 4.0, where potential connections were identified and discussed between the main keywords (by number of co-occurrences) of PSEP with industry 4.0; and (b) the analy- sis of similarities of intersections between PSEP and industry 4.0, where possible gaps were identified by the non-existent intersec- tion between keywords co-occurrences. 5.2. Implications for theory and practice The literature has highlighted the benefits from industry 4.0 concepts and technologies adoption for organizations (Geissbauer et al., 2016), logistics activities (Shang et al. 2013), sustainable oper- ations (Jabbour et al., 2018b), and sustainable concepts (Jabbour et al., 2018a; Jabbour et al., 2017). Angell and Klassen (1999) have argued about the necessity of academic works integrating emerging concepts and technologies of productive systems into environmen- tal issues. However, comprehensive literature reviews in industry 4.0 and PSEP concepts integration have not been addressed. Early research integrating industry 4.0 and sustainable production con- cepts include the works of Jabbour et al. (2018a, 2018b) and Jabbour et al. (2017). Furthermore, the work presents a set of potential connections between PSEP and industry 4.0, as well as existing keywords similarities. There are opportunities to advance knowl- edge into these fields, by integration of industry 4.0 concepts and technologies into the PSEP field. 5.3. Managerial implications This work presents a set of contributions to managers inter- ested in integrating industry 4.0 concepts and technologies into PSEP. The integration between the explored relationships suggests directions of potential integrations between the concepts and tech- nologies of industry 4.0 and PSEP. The proposed matrix as well Envir a p o a a t a w 5 l a t o c i m f u f v a a t A v E f W f A t R A A A A B C C D F F G Nawahda, A., 2016. An assessment of adding value of traffic information and other J.A. Gobbo Junior et al. / Process Safety and s the identification of keywords intersections gives to managers ractical insights on the adoption of the concepts and technologies f industry 4.0 in PSEP fields. This would lead managers toward the doption of strategies for digitalization of operations. The matrix llows managers to foresee potential applications of concepts and echnologies of industry 4.0 in PSEP, considering a micro (factory) nd macro level (productive systems), and examples of applications ere provided that would lead to innovations into these fields. .4. Research limitations The main idea of bibliometric network visualization is to allow arge amounts of complex bibliographic data to be analysed by visu- lizing core aspects of data. Loss of information takes place when extual data is reduced to a co-occurrence network of terms. Loss f information is especially problematic because it may affect the onclusions that can be drawn from a bibliometric network visual- zation. Given the above difficulties, it is useful to visualize a biblio- etric network as a complement to rather than as a substitute or expert judgment. Bibliometric network visualizations are most seful when they are interpreted by experts. Another aspect is that or a small amount of data, there often is no benefit in the use of isualization, because it would be better to directly study the avail- ble data. Qualitative studies are suggested in order to understand doption and integration issues related to identified concepts and echnologies in PSEP fields. cknowledgement This research was supported by UNESP – Sao Paulo State Uni- ersity and CAPES – Coordination for the Improvement of Higher ducation Personnel – PNPD grant. We had received a helpful input rom John Mugge (Proof-Reading-Service.com, Devonshire, UK). e also thank the anonymous reviewers and the editors for their eedback. ppendix A. Supplementary data Supplementary data associated with this article can be found, in he online version, at https://doi.org/10.1016/j.psep.2018.05.017. eferences ngell, L.C., Klassen, R.D., 1999. Integrating environmental issues into the main- stream: an agenda for research in operations management. J. Oper. Manag. 17 (5), 575–598. ntonsen, S., 2009. Safety Culture: Theory, Methods and Improvement. Taylor & Francis, Boca Raton. tzori, L., Iera, A., Morabito, G., 2010. The internet of things: a survey. Comput. Netw. 54 (15), 2787–2805. ven, T., 2016. Risk Assessment and Risk Management: review of recent advances on their foundation. Eur. J. Oper. Res. 253 (1), 1–13. ahr, N., 1997. System Safety Engineering and Risk Assessment: a Practical Approach, first ed. Taylor & Francis Group, Washington. allon, M., Courtial, J.P., Turner, W.A., Bauin, S., 1983. From translations to prob- lematic networks: an introduction to co-word analysis. Soc. Sci. Inf. 22 (2), 191–235. ameron, I., Mannan, S., Németh, E., Park, S., Pasman, H., Rogers, W., Seligmann, B., 2017. Process hazard analysis, hazard identification and scenario definition: are the conventional tools sufficient, or should and can we do much better? Process Saf. Environ. Prot. 110, 53–70. ong, R., Jia, Z., Li, S., 2018. Risk assessment and sources identification of soil heavy metals in a typical county of Chongqing Municipality, Southwest China. Process Saf. Environ. Prot. 113, 275–281. abiano, B., 2017. Editorial: loss prevention and safety promotion in the process industries: issues and challenges. Process Saf. Environ. Prot. 110, 1–4. ortino, G., Guerrieri, A., Lacopo, M., Lucia, M., Russo, W., 2013. An agent-based middleware for cooperating smart objects. Commun. Comput. Inf. Sci., 387–398. eissbauer, R., Vedsø, J., Schrauf, S., 2016. A Strategist’s Guide to Industry 4.0: Global Businesses Are About to Integrate Their Operations into a Seamless Digital Whole, and Thereby Change the World. Strategy + business, pp. 83. onmental Protection 117 (2018) 372–382 381 Gibson, N., 1999. Process safety-A subject for scientific research. Trans. IChemE 77, 153–179. Gobbo Jr., J.A., De Souza, M.G.Z.N., Gobbo, S.C.O., 2017. Barriers and challenges to smart buildings’ concepts and technologies in Brazilian social housing projects. Int. J. Sustain. Real Estate Const. Econ. 1 (1), 31–50. Gubbi, J., Buyya, R., Marusic, S., Palaniswami, M., 2013. Internet of Things (IoT): A vision architectural elements, and future directions. Future Gener. Comput. Syst. 29, 1645–1660. Gunes, V., Peter, S., Givargis, T., Vahid, F., 2014. A survey on concepts, applications and challenges in Cyber-Physical Systems, KSII Trans. Internet Inf. Syst. 8 (12), 4242–4268. Gurjar, B.R., Butler, T.M., Lawrence, M.G., Lelieveld, J., 2008. Evaluation of emissions and air quality in megacities. Atmos. Environ. 42, 1593–1606. Gysin, H., 1976. Environmental-protection and industry. Fresenius Z. Anal. Chem. 282 (4), 389–394. Harzing, A.W., Alakangas, S., 2016. Google Scholar, Scopus and the Web of Science: a longitudinal and cross-disciplinary comparison. Scientometrics 106, 787–804. Haughton, G., Counsell, D., 2004. Regions and sustainable development: regional planning matters. Geogr. J. 170 (2), 135–145. Hauhang, G.Q., Zhong, R.Y., Ysui, K.L., 2015. Special issue on big data for service and manufacturing supply chain management. Int. J. Prod. Econ. 165, 172–173. Heikkilä, A.-M., 1999. Inherent Safety in Process Plant Design—An Index-Based Approach. VTT, Automation, Espoo, Finland (Ph.D. Thesis). Hendershot, D.C., 2006. Lessons from human error incidents in process plant. Process Saf. Environ. Prot. 84, 174–178. Hutton, M., 2017. Making the Link Between Process Safety and Sustainability, https://www.erm.com/en/insights/feature-articles/making-the-link-between- process-safety-and-sustainability/. (Accessed 12 December 2017). Jabbour, C.J.C., Jabbour, A.B.L.S., Sarkis, J., Godinho Filho, M., 2017. Unlocking the circular economy through new business models based on large-scale data: an integrative framework and research agenda. Technol. Forecast. Soc. Change, 1–7 (in press). Jabbour, A.B.L.S., Jabbour, C.J.C., Foropon, C., Godinho Filho, M., 2018a. When titans meet –Can industry 4.0 revolutionise the environmentally sustainable manufac- turing wave? The role of critical success factors. Technol. Forecast. Soc. Change, 1–8 (in press). Jabbour, A.B.L.S., Jabbour, C.J.C., Godinho Filho, M., Roubaud, D., 2018b. Industry 4.0 and the circular economy: a proposed research agenda and original roadmap for sustainable operations. Ann. Oper. Res., 1–16 (in press). Jabbour, C.J.C., 2013. Environmental training in organizations: from a literature review to a framework for future research. Resour. Conserv. Recycl. 74, 144–155. Johnson, D.L., Amborse, S.H., Bassett, T.J., Bowen, M.L., Crummey, D.E., Isaacson, J.S., Johnshon, D.N., Lamb, P., Saul, M., Winter-Nelson, A.E., 1997. Meanings of environmental terms. J. Environ. Qual. 26, 581–589. Kagermann, H., Lukas, W., Wahlster, W., 2011. Industrie 4.0–Mit dem Internet der Dinger auf dem Weg zur 4. Industriellen Revolution. VDI Nachrichten, 13. Kang, H.S., Barczack, G., Nicholas, J., Ledwith, A., Perks, H., 2012. An examination of new product development best practice. J. Prod. Innov. Manag. 29, 180–192. Kehua, S., Li, J., Fu, H., 2011. Smart city and the applications. 2011 International Conference on Electronics Communications and Control (ICECC), 1028–1031. Kesidou, E., Demirel, P., 2012. On the drivers of eco-innovations: empirical evidence from the UK. Res. Policy 41 (5), 862–870. Khan, F., Rathnayaka, S., Ahmed, S., 2015. Methods and models in process safety and risk management Past, present and future. Process Saf. Environ. Prot. 98, 116–147. Kraft, M.E., 2017. Environmental Policy and Politics. Routledge, New York. Lee, I., Lee, K., 2015. The Internet of Things (IoT) Applications, investments and challenges for enterprises. Bus. Horiz. 58, 431–440. Lee, J., Kao, H., Yang, S., 2014. Service innovation and smart analytics for industry 4.0 and big data environment. Procedia CIRP 16, 3–8. Lee, J., Bagheri, B., Kao, H.A., 2015. A cyber-physical systems architecture for industry 4.0-based manufacturing systems. Manufac. Lett. 3, 18–23. Li, Z., Zongwei, M., Kuijp, T.J., Yuan, Z., Huang, L., 2014. A review of soil heavy metal pollution from mines in China: pollution and health risk assessment. Sci. Total Environ. 468–469, 843–853. Lu, S.Q., Xie, G., Chen, Z., Han, X., 2015. The management of application of big data in Internet of Thing in Environmental Protection in China. IEEE First International Conference on Big Data Computing Service and Applications. Luiz, J.V.R., Jugend, D., Jabbour, C.J.C., Luiz, O.R., De Souza, F.B., 2016. Ecodesign field of research throughout the world: mapping the territory by using an evolutionary lens. Scientometrics 109, 241–259. Mazurkiewicz, P., 2004. Corporate environmental responsibility: is a common CSR framework possible? In: 24th Annual IAIA Conference, Vancouver, BC. Mishra, S.A., Tijare, D.S., Asutkar, G.M., 2011. Design of energy aware air pollution monitoring system using WSN. Int. J. Adv. Eng. Tech 1 (2), 107–116. Mittal, S., Khan, M.A., Romero, D., Wuest, T., 2017. Smart manufacturing: charac- teristics, technologies and enabling factors. Proc IMechE Part B: J Eng. Man., 1–20. Mongeon, P., Paul-Hus, A., 2016. The journal coverage of Web of Science and Scopus: a comparative analysis. Scientometrics 106, 213–228. attributes as part of its classifiers in a data mining tool set for predicting surface ozone levels. Process Saf. Envir. Prot. 99, 149–158. OSHA, 2000. Process Safety Management. U.S. Department of Labor Occupational Safety and Health Administration, OSHA 3132 2000 (Reprinted). https://doi.org/10.1016/j.psep.2018.05.017 https://doi.org/10.1016/j.psep.2018.05.017 https://doi.org/10.1016/j.psep.2018.05.017 https://doi.org/10.1016/j.psep.2018.05.017 https://doi.org/10.1016/j.psep.2018.05.017 https://doi.org/10.1016/j.psep.2018.05.017 https://doi.org/10.1016/j.psep.2018.05.017 https://doi.org/10.1016/j.psep.2018.05.017 https://doi.org/10.1016/j.psep.2018.05.017 https://doi.org/10.1016/j.psep.2018.05.017 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0005 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0005 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0005 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0005 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0005 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0005 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0005 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0005 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0005 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0005 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0005 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0005 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0005 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0005 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0005 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0005 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0005 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0005 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0005 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0005 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0005 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0005 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0010 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0010 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0010 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0010 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0010 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0010 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0010 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0010 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0010 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0010 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0010 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0015 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0015 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0015 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0015 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0015 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0015 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0015 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0015 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0015 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0015 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0015 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0015 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0015 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0020 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0020 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0020 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0020 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0020 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0020 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0020 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0020 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0020 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0020 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0020 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0020 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0020 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0020 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0020 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0020 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0020 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0020 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0020 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0020 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0020 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0025 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0025 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0025 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0025 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0025 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0025 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0025 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0025 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0025 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0025 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0025 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0025 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0025 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0025 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0025 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0025 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0030 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0030 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0030 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0030 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0030 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0030 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0030 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0030 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0030 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0030 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0030 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0030 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0030 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0030 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0030 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0030 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0030 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0030 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0030 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0035 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0040 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0040 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0040 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0040 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0040 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0040 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0040 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0040 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0040 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0040 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0040 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0040 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0040 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0040 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0040 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0040 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0040 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0040 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0040 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0040 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0040 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0040 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0040 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0040 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0040 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0040 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0045 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0045 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0045 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0045 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0045 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0045 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0045 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0045 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0045 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0045 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0045 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0045 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0045 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0045 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0045 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0045 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0045 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0045 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0045 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0045 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0045 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0050 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0050 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0050 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0050 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0050 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0050 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0050 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0050 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0050 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0050 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0050 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0050 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0050 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0050 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0055 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0060 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0060 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0060 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0060 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0060 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0060 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0060 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0060 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0060 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0060 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0060 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0060 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0065 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0065 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0065 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0065 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0065 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0065 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0065 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0065 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0065 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0065 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0065 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0065 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0065 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0065 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0065 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0065 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0065 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0065 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0065 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0065 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0065 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0065 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0065 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0065 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0065 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0065 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0070 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0070 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0070 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0070 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0070 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0070 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0070 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0070 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0070 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0070 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0070 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0070 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0070 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0070 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0070 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0070 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0070 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0070 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0070 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0075 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0075 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0075 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0075 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0075 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0075 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0075 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0075 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0075 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0075 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0075 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0075 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0075 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0075 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0075 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0075 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0075 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0075 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0075 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0075 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0080 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0080 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0080 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0080 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0080 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0080 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0080 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0080 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0080 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0080 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0080 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0080 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0080 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0080 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0085 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0085 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0085 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0085 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0085 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0085 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0085 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0085 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0085 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0085 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0085 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0085 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0090 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0090 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0090 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0090 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0090 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0090 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0090 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0090 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0090 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0090 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0090 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0090 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0090 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0090 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0090 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0090 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0090 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0090 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0095 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0095 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0095 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0095 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0095 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0095 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0095 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0095 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0095 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0095 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0095 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0095 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0095 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0095 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0100 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0100 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0100 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0100 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0100 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0100 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0100 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0100 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0100 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0100 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0100 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0100 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0100 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0100 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0100 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0100 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0100 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0100 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0100 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0100 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0105 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0105 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0105 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0105 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0105 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0105 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0105 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0105 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0105 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0105 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0105 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0105 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0105 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0105 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0105 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0105 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0110 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0110 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0110 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0110 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0110 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0110 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0110 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0110 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0110 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0110 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0110 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0110 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0110 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0110 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0110 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0110 https://www.erm.com/en/insights/feature-articles/making-the-link-between-process-safety-and-sustainability/ https://www.erm.com/en/insights/feature-articles/making-the-link-between-process-safety-and-sustainability/ https://www.erm.com/en/insights/feature-articles/making-the-link-between-process-safety-and-sustainability/ https://www.erm.com/en/insights/feature-articles/making-the-link-between-process-safety-and-sustainability/ https://www.erm.com/en/insights/feature-articles/making-the-link-between-process-safety-and-sustainability/ https://www.erm.com/en/insights/feature-articles/making-the-link-between-process-safety-and-sustainability/ https://www.erm.com/en/insights/feature-articles/making-the-link-between-process-safety-and-sustainability/ https://www.erm.com/en/insights/feature-articles/making-the-link-between-process-safety-and-sustainability/ https://www.erm.com/en/insights/feature-articles/making-the-link-between-process-safety-and-sustainability/ https://www.erm.com/en/insights/feature-articles/making-the-link-between-process-safety-and-sustainability/ https://www.erm.com/en/insights/feature-articles/making-the-link-between-process-safety-and-sustainability/ https://www.erm.com/en/insights/feature-articles/making-the-link-between-process-safety-and-sustainability/ https://www.erm.com/en/insights/feature-articles/making-the-link-between-process-safety-and-sustainability/ https://www.erm.com/en/insights/feature-articles/making-the-link-between-process-safety-and-sustainability/ https://www.erm.com/en/insights/feature-articles/making-the-link-between-process-safety-and-sustainability/ https://www.erm.com/en/insights/feature-articles/making-the-link-between-process-safety-and-sustainability/ https://www.erm.com/en/insights/feature-articles/making-the-link-between-process-safety-and-sustainability/ http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0120 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0120 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0120 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0120 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0120 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0120 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0120 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0120 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0120 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0120 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0120 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0120 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0120 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0120 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0120 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0120 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0120 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0120 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0120 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0120 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0120 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0120 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0120 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0120 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0120 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0120 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0120 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0125 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0130 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0130 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0130 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0130 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0130 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0130 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0130 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0130 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0130 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0130 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0130 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0130 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0130 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0130 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0130 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0130 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0130 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0130 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0130 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0130 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0130 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0130 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0130 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0130 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0135 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0135 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0135 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0135 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0135 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0135 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0135 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0135 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0135 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0135 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0135 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0135 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0135 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0135 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0135 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0135 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0135 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0135 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0135 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0135 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0135 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0140 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0140 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0140 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0140 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0140 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0140 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0140 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0140 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0140 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0140 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0140 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0145 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0145 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0145 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0145 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0145 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0145 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0145 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0145 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0145 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0145 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0145 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0145 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0145 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0145 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0145 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0145 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0145 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0145 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0150 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0150 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0150 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0150 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0150 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0150 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0150 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0150 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0150 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0150 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0150 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0150 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0150 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0150 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0150 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0150 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0155 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0155 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0155 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0155 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0155 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0155 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0155 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0155 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0155 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0155 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0155 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0155 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0155 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0155 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0155 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0155 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0155 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0160 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0160 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0160 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0160 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0160 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0160 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0160 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0160 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0160 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0160 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0160 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0160 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0160 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0160 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0160 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0160 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0160 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0165 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0165 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0165 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0165 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0165 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0165 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0165 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0165 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0165 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0165 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0165 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0165 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0165 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0165 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0165 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0165 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0165 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0165 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0165 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0165 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0165 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0170 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0170 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0170 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0170 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0170 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0170 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0170 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0175 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0175 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0175 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0175 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0175 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0175 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0175 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0175 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0175 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0175 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0175 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0175 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0175 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0175 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0175 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0175 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0175 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0180 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0180 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0180 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0180 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0180 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0180 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0180 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0180 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0180 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0180 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0180 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0180 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0180 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0180 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0180 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0180 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0180 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0180 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0185 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0185 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0185 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0185 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0185 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0185 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0185 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0185 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0185 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0185 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0185 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0185 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0185 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0185 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0185 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0190 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0190 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0190 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0190 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0190 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0190 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0190 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0190 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0190 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0190 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0190 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0190 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0190 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0190 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0190 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0190 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0190 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0190 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0190 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0190 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0190 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0190 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0190 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0190 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0190 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0195 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0195 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0195 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0195 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0195 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0195 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0195 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0195 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0195 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0195 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0195 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0195 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0195 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0195 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0195 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0195 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0195 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0195 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0195 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0195 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0195 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0195 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0195 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0195 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0195 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0195 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0195 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0200 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0200 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0200 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0200 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0200 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0200 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0200 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0200 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0200 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0200 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0200 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0200 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0200 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0200 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0200 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0200 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0200 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0200 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0200 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0200 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0205 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0205 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0205 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0205 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0205 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0205 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0205 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0205 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0205 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0205 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0205 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0205 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0205 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0205 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0205 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0205 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0210 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0210 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0210 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0210 http://refhub.elsevier.com/S0957-5820(18)30191-5/sbref0210 http://refhub.elsevier.com/S0957-5