THE AUTOMATION OF INDUSTRIAL PLANTS AND THE EPC CONTRACTS

IBP1025_11 THE AUTOMATION OF INDUSTRIAL PLANTS AND THE EPC CONTRACTS (ENGINEERING, PROCUREMENT AND CONSTRUCTION) 1 2 3 Carlos Eduardo R. Barateiro , Carlos H. Caldas , José Rodrigues de Farias Filho , 4 Luiz Antonio da Paz Campagnac Copyright 2011, Brazilian Petroleum, Gas and Biofuels Institute - IBP This Technical Paper was prepared for presentation at the Rio Pipeline Conference & Exposition 2011, held between September, 20-22, 2011, in Rio de Janeiro. This Technical Paper was selected for presentation by the Technical Committee of the event. The material as it is presented does not necessarily represent Brazilian Petroleum, Gas and Biofuels Institute’ opinion or that of its Members or Representatives. Authors consent to the publication of this Technical Paper in the Rio Pipeline Conference & Exposition 2011. Abstract This article presents the results of a qualitative research conducted with directors, managers and consultants of oil and gas end users, companies specialized in EPC contracts, suppliers and integrators of automation systems - all of them operating in the country. The main objective of investigation is to determine how the implementation of automation should be hired for O&G large projects. As secondary objective, to understand why the traditional structures are not maximizing the gains of digital plants concept; to demonstrate that for this to happen it is necessary that the automation be hired in the early stages of projects; to analyze the impact of existing hiring models and to investigate the successful critical factors in selecting the best provider of the automation system. 1. Introduction There is an excessive focus on technical specification of the projects and there are few who truly understand that the new technologies available for automation of industrial processes control must change the way businesses are run. The automation projects are becoming more sophisticated with the increasing use of digital plants concept. The benefits of new technologies can be significant in terms of both investment costs (CAPEX) and in operation costs (OPEX), if the execution procedures are also accompanying this development. The dissertation developed has considered the automation evolution in recent decades. Such research culminated in the state of the art available today which are the latest versions of DCS's (Distributed Control Systems), called hybrid systems, designed to be used with modern digital networks. With this focus, the work examined different forms of hiring for major projects, especially EPC contracts applied in the O&G areas. On this basis, we have studied the feasibility of using the MAC model (Main Automation Contractor) in which one company is responsible for the automation of all the industrial plant, including materials, software and services inherent to these systems. The study also found that the automation procurement in the early stages of the project could be a solution to the maximization of profits with the digital plant concept. The strategy of anticipating the specific items is advocated by Vorster et al. (1998) of Construction Industry Institute of the University of Texas in Austin (USA), and our study indentified that automation can be a strategic item for projects. The complete dissertation had the following steps: a) Search of the conceptual framework in order to base the instruments and concepts of the study; b) Preparation of the first versions of the survey instruments; c) Selection of respondents for the application of survey instruments; d) Conduction of individual interviews and application of pre-test questionnaires in order to criticize the instruments; e) Transcription, content analysis and comments on interviews and pre-test questionnaires; f) Preparation of final survey instruments; g) Data collection; ______________________________ 1 M.SC., Mechanical Engineer – UNIVERSIDADE FEDERAL FLUMINENSE 2 PHD, Civil Engineer – UNIVERSITY OF TEXAS AT AUSTIN 3 D.SC, Civil Engineer – UNIVERSIDADE FEDERAL FLUMINENSE 4 D.SC, Civil Engineer – UNIVERSIDADE FEDERAL FLUMINENSE Rio Pipeline Conference & Exposition 2011 h) Transcription of interviews and tabulation of quantitative data; i) Implementation of quantitative analysis; j) Implementation of content analysis; k) Preparation of final report. This article reports the results of this field research, obtained through personal interviews with directors, managers and consultants of O&G market including end users, EPC contractors, automation vendors and system integrators that have operations in Brazil. We used a standardized and structured questionnaire to get the viewpoint in terms of processes, people and technology. 2. The Field Research We conducted twenty-five individual interviews that was the basis for analysis content, which is a technique that can be used to obtain information expressed in verbal or written form, summarized and organized as defined by Freitas (2000). We had the following distribution for the interviewees (Figure 1): Figure 1: Distribution of Interviewees by Segment (2010) 3. Survey Construction The case study was conducted in an exploratory way as defined by Yin (2001) and a survey was developed with twenty-two questions designed to allow responses to the four main subjects: Q1) why do not the current structures maximize the gains of digital plants? Q2) could the hiring of automation bring impact to the project implementation? Q3) could the hiring of automation in the early stages of projects, provide gains to the enterprise? Q4) what are the successful critical factors for a hiring model that maximize the gains of digital plants? These subjects were responded and have led to conclusions considering the aspects of the processes, people and technology. The interviews were conducted through a semi-structured instrument as defined by Yin (2001). Such instrument based the quantitative and qualitative analysis of perception. The research method is therefore qualitative and quantitative. "While the quantitative method measures the object, the qualitative method measures the categories and their attributes" as defined by Oliveira (2000). 4. Interview Analysis . 4.1 Answering Q1 The first big question had the objective of identifying the reasons why the current structures do not maximize gains on digital plants. For that answer, six questions (P1 to P6) were formulated to the participants. 4.1.1 Analysis of P1 This first question searched the interviewee’s position whether the automation had created impacts in the projects implementation already undertaken and what kind of impacts had occurred. Most of the responses (92%) confirmed that there were significant impacts mainly because automation is the "heart" of the plants. It directly affects the operational expenditure (OPEX). There were positive and negative impacts depending on each project, in terms of price, time and resources. 2 Rio Pipeline Conference & Exposition 2011 Notably, most of impacts are more the result of delays in carrying out activities related to automation and therefore, with impact in project start up. The automation impact cost itself is small because the automation represents very little compared to the values of other parts of the project. However, its impact on the project cost is indirect: the delay in the project start up is what may represent the most significant financial aspect. In terms of resources, the biggest impact is the fact that as the automation is triggered at the end of the project. So, it is necessary the mobilization of teams, greater than expected, to compensate the delays in the entire project. Many positive aspects were also mentioned, mainly when the automation brought the increase of efficiency, productivity and results in the industrial process in general. Despite the fact that the evolution of automation technology has an increasing price, it was reported a reduction in price per point of I/O (input/output) over the years. There were many who reported that the main cause of negative impacts is directly related to the planning aspects of development: as the automation has a low cost within the project budgets, it is given little importance to it, then it is only hired in the implementation of final stages. Other interviewees also described that the lack of a greater involvement among the contractor, the end user and the supplier of the system, since the early stages of the project, is an important cause for these impacts: because it ends up causing miscommunication between the automation and process discipline. There were also some reports which showed that the lack of experienced staff on the market, especially in the automation discipline, causes problems to the project quality. The best experiences mentioned happened when this involvement occurred since the early stages of the development and when the definitions of scope were clearly defined. 4.1.2 Analysis of P2 This second question searched the interviewee’s position whether the project team was sufficiently updated with the latest technologies. There was almost unanimity (88%) in the responses: the lack of updating is mainly caused because the development of automation technologies is growing nearly at the same rate as information technology. Therefore it is very difficult to keep teams properly updated because there is no time available for completion of the required training. It was also mentioned another important aspect: there are even many professionals with good academic background and excellent technical aspects, but with very little field experience which even affects the use of not so recent technologies. It was mentioned that such lack of experience is causing the lack of use of various resources available in the automation systems. It happens because the automation suppliers do not know the practical applications and because of this, they cannot identify what the end user really needs which is usually much less than what is available in systems. 4.1.3 Analysis of P3 This third question searched the interviewee’s position whether the project team dominated all the features that are part of an automation system. There was unanimity (92%) in this item because it is very rare to find a staff with a broad knowledge of all these features. What is found on the market are specialists in certain areas, which is understandable because of the complexity of automation. For the implementation of the projects, the great difficulty is to find leaders who have an understanding of the whole and who are able to coordinate these multidisciplinary teams. 4.1.5 Analysis of P4 This forth question searched the interviewee’s position about the involvement of the responsible for the automation project on the detailed project development. We had almost unanimity (80%): typically the suppliers deliver what was hired. The basic reason for that is the hiring model used in Brazil: focused on the lowest price. It leads the suppliers to prepare their proposals to meet the minimum specifications (minimum compliance). So, during the development of projects, the suppliers´ commercial interface is much more active than their own technical interface, limiting the implementation of improvements and even jeopardizing the longevity and scalability of the implemented solutions. Another aspect mentioned was the moment when the automation is hired, much more in the final stages of the projects. The biggest cause of this limitation is that any changes have costs and impacts in the delivery time, which normally are not accepted by both the contractor and the end user. The best experiences were reported when the hiring model, defined by the contractor for the automation system, created the conditions for this engagement to take place and when the automation was hired in the early stages of the engineering project development. 4.1.5 Analysis of P5 3 Rio Pipeline Conference & Exposition 2011 This fifth question searched the interviewee’s position whether the project implementation is using all the benefits that the digital plant concept can provide. We also had almost unanimity (84%): all possible advantages are not being used, especially because the people, who carry out the project, still make it in a too conservative way, due to the lack of knowledge in new technologies and lack of applicative experience. It was mentioned that perhaps the biggest gains in the use of the digital plant concept could occur in the commissioning and start up, taking advantage of the available intelligence on the field. However, this is not being used because the procedures of these activities were not adapted to the new technologies and there is a great lack of information in the use of these tools. 4.1.7 Analysis of P6 This sixth question searched the interviewee’s position whether the latest development projects and in operation are using all the advantages of new technologies. We also had almost unanimity (80%): what is in use is very little of these benefits. Several reasons were reported, but the main aspect was the distance between the teams responsible for design and the teams responsible for the enterprise operation. It was mentioned that the plants are evolving and new enhancements are being applied, but the projects do not have the focus on who will operate, in other words, without considering the operator of “the real world”. Moreover, huge gaps were also mentioned in the training for operators, which is typically in the end of the project completion. It relegates such important activity to a second plan, not allowing the operators to be fully aware of the procedures that must be performed. It was mentioned that the turnover among the people from the operation staff also contributes greatly to the non-use of these benefits. Another aspect was also mentioned: the lack of integration among the various parts of the automation systems, field instruments, control room, applicative software and communication links, eventually does not allow the information flow to be appropriate. It was reported some good experiences in refining area, petrochemical area and mainly in the pulp and paper area. 4.1.7 Conclusion of Q1 Analyzing the positions found for the six questions we can answer why the current structures do not maximize the gains of the digital plants based on the aspects of processes, people and technology: a) Process: The current hiring model based on the lowest price entails the delivery of the minimum necessary to meet the specifications. Given the very low importance to the budget in relation to the other project values, the automation is left to a second plan. So, it is hired in the final stages of the projects, limiting the implementation of improvements. This leads to a little involvement of end users, contractors and suppliers in the development of projects; b) People: The speed of automation technological evolution is as fast as the information technology, making it very difficult to let the teams trained. There is also a lack of field experience for the personnel who are involved in project execution. There is a shortage of generalist professionals in all features of automation, hindering the coordination of multidisciplinary teams needed for the development of automation projects. c) Technology: There is still an excessive conservatism in the project implementation and the procedures are not updated with the technology advances. This technology used is also not considering the real world of the operators due to lack of applicative experience in the project execution. There is also a lack of proper integration (communication) among the various parts that make up the technological packages for an automation system. 4.2 Answering Q2 The second big question had the objective of identifying the impacts caused by hiring model in the implementation of automation projects. For that answer, six questions (P7 to P12) were formulated to the participants. 4.2.1 Analysis of P7 This first question searched the interviewee’s position about of the current form for hiring of the automation system vendor. It was observed that the vast majority answered (72%) that there is a trending of the hiring to be by general register of pre-qualified vendors or specific vendor list to certain projects. The hiring option by general register is more applicable when the focus is the purchase of hardware. The second alternative is more used for expansions of existing projects or when you want to set a pattern on something already known. It was observed that the hiring cases through partnership agreements were what generated the best results, but it has also been reported that this model is difficult to implement in state companies due to mechanisms and legal rules. It was 4 Rio Pipeline Conference & Exposition 2011 also mentioned that in other countries this kind of hiring is very common. But in Brazil there are many suspicions among end users that the suppliers take advantage of the partnership to enhance their prices and not to establish a win-win partnership. 4.2.1 Analysis of P8 This second question searched the interviewee’s position whether the hiring model called MAC, in which a single supplier is the responsible for all automation, would bring gains to the large projects. We had almost unanimity (92%) among all the interviewees that this model brings gains mainly due to the reduction of interfaces, but some important aspects were mentioned: - There must be integration between automation and electrical package: or it must be done by the same supplier or one company must be responsible for the integration of both packages; - The MAC is recommended for new plants (new field) and not for existing projects because in such cases, its implementation produces fewer effects; - The MAC benefits are clearer when we have a large project with many EPC contractors because, in such case, the MAC can standardize the automation among all of them; - The MAC concept is much more than simply to supply instruments, hardware, software and inherent automation services – it must be part of the project development for the advantages to be visible; - The use of this model can be considered a serious risk to Brazilian integrators that are likely to lose ground in the implementation of large projects for multinational companies. It was also featured a very interesting aspect about the gains of the MAC: it is very difficult, in the beginning of a project, to affirm categorically that there are gains with this model. The reduction of interfaces is much more a feeling of everyone than actually something tangible. There were even comments that have reported serious relationship problems among the contractors, the end users and the MAC providers during the development of some projects notably in terms of commercial aspects. 4.2.2 Analysis of P9 This third question searched the interviewee’s position whether the MAC should include instruments and field equipment. This was perhaps the most controversial item of the survey. Most respondents (68%) confirmed that the MAC supplier shall be responsible for the whole package: due to the reduction of interfaces and to allow better integration among the control room, applicative packages and field equipment. But we observed some important comments on the part of valves and intrusive line components. Due to the size of these packages, their technical features and impact on projects of pipelines, many believe these items should stay with the contractor. Regardless, there is consensus that in such cases, the MAC provider must have active participation in the purchasing management of these items. There were also comments that the inclusion of the field equipment in the MAC scope could limit the competition among the companies because there are few that can supply a complete package. 4.2.3 Analysis of P10 This fourth question searched the interviewee’s position on who should hire the MAC. This item also had a serious different position among the considered segments. For contractors, there is a clear position (four opinions in five answers) in favor that the hiring of MAC should be their responsibility. And this position is based on the following: a) The EPC definition is clear: there is a unique responsible for the project. So, the MAC, which is an interface between the end user and contractor, is not appropriate for this kind of project; b) The interfaces between the end user and the contractor, considering interference of third parts, have always a very tough resolution due to the various commercial aspects, notably by the detachment of costs and revenues. For the remaining segments (end users, manufacturers, integrators and opinion leaders), the following reasons were mentioned for the MAC hiring by the end user (eighteen opinions in twenty answers): c) The end user has a concern with the operation of the plant (long term) and the contractor with the delivery of the project at the lowest possible cost (short term); d) The end user knows the process technology and can, interfacing with the MAC supplier, help in the developing of a more adequate system to its needs; e) Automation has little relevance to the contractor. So, it is not given the proper importance to its implementation. It was also mentioned an alternative to this problem: the MAC hiring by contractor based on specification elaborated by the end user. However, this is not always possible because the end user needs to be updated with the automation technology and to have an engineering structure, which is not always available because it is not its core business. 5 Rio Pipeline Conference & Exposition 2011 4.2.4 Analysis of P11 This forth question searched the interviewee’s position whether teams who are responsible for the project implementation would like to have contact with the MAC supplier. We had almost unanimity (80%) in the answers: it would be very positive because there would be convergence of goals and productivity gains. It was said that this strategy would allow the optimization of processes and the simplification of documentation and its flow. We received several comments that perhaps there could be some barriers for the implementation by the project team, because this is not a common used practice in projects already developed. 4.2.5 Analysis of P12 This fifth question searched the interviewee’s position whether the existing features in automation systems justify the MAC hiring. Most responses (72%) were positive based on today´s existing complexity in automation systems and the lack of qualified labor in the market. There were also many comments that only one MAC type contract could implement many of the features available in current technology. The opposite opinions noted that the MAC is a concept of project execution and is not related to the features that will be implemented, once there was little performance difference among the different vendors of systems. 4.2.6 Conclusion of Q2 Analyzing the positions found for the questions we considered the impacts caused by the hiring model in the implementation of projects, based on aspects of processes, people and technology: a) Process: The current hiring model is based on general register and vendor list and there is still little engagement through partnerships, especially in the public sector, despite the many reports of success with this kind of model. There are a great number of benefits in the use of the MAC concept, in which a single supplier is responsible for automation, mainly due to reduction of interfaces. This model is recommended for new plants, it must provide the integration with the electrical packages and the benefits are clearer when we have a large project with many EPC contractors because, in such case, the MAC can standardize the automation among all of them. The MAC concept is much more than simply to supply instruments, hardware, software and inherent automation services – it must be part of the project development for the advantages to be visible. The answers informed us that the MAC should consider the full scope of automation, including the field part. We had many doubts if intrusive equipments should belong to that package, but in this case, all comments were that the purchasing management of these intrusive items must have the participation of MAC provider. There is a trend that the MAC hiring should be made by end user because it is aware of the process. But it should be considered possible interface problems with the contractor; b) People: There is a consensus that the personnel involved in the project would be interested in working with the MAC supplier because it would allow the simplification of project activities; c) Technology: Most believe that the current existing features justify the MAC hiring due to the complexity of automation systems and the lack of qualified labor. 4.3 Answering Q3 The third big question had the objective of identifying whether the hiring of automation in the early stages of the projects would provide gains for the enterprise. For that answer, five questions (P13 to P17) were formulated to the participants. 4.4 Analysis of P13 This first question searched the interviewee’s position whether there are gains in PEpC model, in which certain strategic items are purchased in advance, most of the times even before the very definition of EPC contractor. The majority responded (64%) affirmatively noting that this type of purchase provides gains when this item is important: through its impact on the project schedule or because it is a source of information for the project development. It was also mentioned that the highest expenses in the projects are not exactly in the hardware cost (bigger impact in the beginning), but the amount of man-hours spent on the entire project development. Thus this strategic purchasing could help reduce these expenses through the increasing of previous definitions. However, despite positive opinions in most cases, it has been reported several problems that are occurring in Brazil because such purchases were anticipated. Delays in delivery of these items and errors in the sizing of the purchased equipment have caused serious difficulties in the work of contractors. 6 Rio Pipeline Conference & Exposition 2011 4.5 Analysis of P14 This second question searched the interviewee’s position about the moment his company is hiring the automation system vendor. There were many different positions, but most of the cases (76%) showed a trend of hiring be accomplished within the detailed design. There were several comments that the fact of automation being performed in the final stage does not allow much to be done in terms of project improvement. Several end users commented that the automation is not been hired in earlier phases (conceptual) because in this moment there is a concern with the oil and gas production processes and the automation is not the priority. 4.6 Analysis of P15 This third question searched the interviewee’s position whether the automation should be one of the strategic items to be hired in advance as recommended by the model PEpC. Most of the responses (76%) were favorable to it mentioning, not the initial investment (Capex), but the automation importance in the cycle of plant life, due to the fact that it will influence the philosophy of operation and maintenance. This question had a clear position disagreement among the contractors and the others interviewees. There is a concern that such advanced hiring will lead additional efforts, due to renegotiations and the permanence of project teams longer than expected. 4.7 Analysis of P16 This fourth question searched the interviewee’s position whether the teams responsible for the project would like to work with the automation provider since the early stages of project development. There was almost unanimity (88%) that there will be gains in the reduction of uncertainties in the project development. However an important aspect was mentioned: Brazilian laws do not accept that a company that has participated in the basic design or FEED can participate in the end phase of EPC. It leads the big contractors to put the focus on this last phase which has a higher value. 4.8 Analysis of P17 This fifth question searched the interviewee’s position whether the existing features justify the hiring of automation in the early stages of projects. Most confirmed (72%) that it is correct, justified by the complexity in existing automation systems and the lack of qualified labor for it. 4.9 Conclusion of Q3 Analyzing the positions found for the above questions, we can conclude that the hiring of automation in the early stages of the projects would provide gains for the enterprises, based on aspects of processes, people and technology: a) Processes: There are gains in the strategy of the advanced purchase (PEpC), especially if the item is an important source of information for project development. But what is being noted in ongoing projects in Brazil are delays in the delivery of items purchased by end users, who are jeopardizing the implementation of projects. Currently, the automation is been hired in the detailed design, which ends up limiting the implementation of improvements. There is a clear acceptance that automation should be seen as a strategic item and purchased in advance. b) People: There is a consensus that there would be a very large acceptance of the teams responsible for the project implementation in working with the automation supplier. The main reason for this is the reduction of project uncertainty that would be obtained. c) Technology: Most believe that the existing functionalities in automation systems justify their engagement in the early stages of project development. 4.4 Answering Q4 The fourth big question had the objective of identifying the successful critical factors for the hiring model that maximize the gains of digital plants. For that answer, five questions (P18 to P22) were formulated to the participants. 4.4.1 Analysis of P18 7 Rio Pipeline Conference & Exposition 2011 This first question searched the interviewee’s position whether the creation of an automation hiring model based on four stages: definition of scope, preparation of quotations, quotation on the market and criteria for the choice of provider - would be appropriate to define the best solution to purchase an automation system. We had almost unanimity (92%) that these four steps would be appropriate for this choice. There was a comment that maybe, for large projects, an audit to verify the qualifications submitted by the vendor would be important. 4.4.1 Analysis of P19 This second question searched the interviewee’s position whether the technical analysis and technology of the vendor justify a higher price for the hiring. We had practically unanimity (84%) that it is important and that the end user and contractor must get the best price and not the lowest price. For this, it is necessary that the best cost of ownership be considered and not only the cost of initial investment. Despite this general opinion, the common practice of the Brazilian market is that price is the most valuable item. 4.4.2 Analysis of P20 This third question searched the interviewee’s position whether a standard hiring model would be beneficial for the choice of the best solution for the automation provider. The majority confirmed (84%) that it would be interesting to adopt a standard model because it would give impartiality on the process of the supplier´s choice, allowing the buyer to focus on more important activities (negotiation and contract terms), and not to spend time in the creation of evaluation concepts. The opposite views about it emphasized that the use of a standard model could halt the process of choosing these automation suppliers. They defended that it could be harmful because the automation is very dynamic. Another negative aspect raised is that a standard systematic could limit the amount of market participants that certainly would be unable to meet all demands. 4.4.3 Analysis of P21 This fourth question searched the interviewee’s position whether the staff responsible for procurement should have any special qualifications or training to acquire the automation systems. The responses (64%) showed that the procurement team needs a specific qualification for automation. However, we received several comments that, for large projects, the expertise of the engineering sector is always used. Thus the procurement team does not need to have any specific qualification in this area. 4.4.4 Analysis of P22 This fifth question searched the interviewee’s position regarding the degree of importance that certain items should have in the evaluation of a supplier, in other words, the successful critical factors for evaluation. The figure below shows the successful critical factors by importance degree expressed by the interviewees (Figure 2). Figure 2: Importance of Successful Critical Factors (2010) We received some very interesting comments about the successful critical factors: a) Some have suggested that there should be an analysis of the ability of the MAC providers to provide appropriate automation training tailored according to the needs of Brazilian operators; 8 Rio Pipeline Conference & Exposition 2011 b) Others suggested that it is necessary to verify if the supplier is able to continue providing technical support on site after the completion of the project development; c) Others suggested that the project portfolio of the suppliers must define the teams that were responsible for its implementation. Because the fact that the company has executed a particular project is no guarantee that the next will have the same success without the same participants; d) For the same reason some considered important that the presentation of the talent retaining policy be done by the vendor. 4.4.5 Conclusion of Q4 Analyzing the positions for the questions we found the successful critical factors for hiring model that maximizes the gains of digital plans, based on aspects of processes, people and technology: a) Process: The creation of a model based on four stages (scope definition, preparation of quotations, quotation in the market and criteria of analysis) was considered adequate for the choice of the best vendor. It was considered that the technical analysis and technology of a vendor are important points to define the best price despite the fact that this kind of analysis is not practiced in Brazilian market, as a rule. It was considered that a standard hiring model would be beneficial to the market due to exemption in the choice and to provide the correct analysis of the parameters; b) People: It was identified the need of the procurement teams to have, in its staff, people with specialization in automation. But we received many comments showing that for the large projects, the procurement team has the support of the engineering team. In this case, it is not necessary a special qualification for the procurement team. c) Technology: It was quantified, based on the responses, the importance of successful critical factors for choice of the automation provider. 5. Final Conclusion The study showed the importance of automation within the enterprises particularly related to aspects of the complete project life cycle, including its operation. It was demonstrated the necessity and importance that the automation be hired in the early stages of projects for the benefits of the newer technologies to be provide. It was also demonstrated that the hiring of automation using the MAC model is most suitable for large projects, in which a single supplier is responsible for design and supply of all automation systems. 6. References FREITAS, H. M. R.de; JANISSEK, R. Lexical analysis and content analysis: complementary techniques, sequential and applicants for exploration of qualitative data. Porto Alegre: Sagra Luzzatto, 2000. FURLAN, J. D.; IVO, I. M.; AMARAL, F. P. Executive Information Systems. São Paulo: Makron Books, 1994 YIN, R. K., Case Study: planning and methods. Porto Alegre; Bookman, 2001. OLIVEIRA, C. S., Scientific Methodology, planning and survey techniques. São Paulo: LTr, 2000 ROCKART, J. F. , A Primer on Critical Success Factors, Harvard Business Review, p. 81-93, mar. /abr. 1979 VORSTER, Michael C.; MAGROGAN, Stephanie A.; MCNEIL, Blair W... A Breakthrough Project Delivery System that Improves Performance by Reforming Owner, Contractor, Supplier Relationships: Under the Guidance of the of the Owner/Contractor/Supplier Relationships Research Team Number 130. First edition Austin: Construction Industry Institute, 1998. 324 p 9