| HOME Software Testing Data Mining and Warehousing |
Basics: This unit covers project implementation and management for process control systems. It should not be taken "as is" because statutory, technical, and corporate needs vary from one project to the other and from one site to the other. Also, this unit cannot cover all possibilities because every project is different-however, it will attempt to take a start-to-finish approach. All projects require careful planning, particularly when control systems are complex and implemented with tight budgets and short schedules. For all projects, documents are vital. Reference documents should show how, why, and when a decision was reached and by whom. This makes it easier for others to pick up later and provides all the reasons and history behind technical decisions. Project personnel typically consist of a client, engineering personnel, equipment suppliers, and contractors. The client's work should be clearly defined through documents (e.g., specifications and drawings). The project manager must coordinate the activities of the client, engineering personnel, suppliers, and contractors. Engineering personnel, suppliers, and contractors should conform to the client's requirements (as identified in all the documents produced) and be in compliance with the required codes and standards. Generally, a project starts because of the following: • potential market opportunities, or • disposal (or conversion) of a process by-product is required, or • compliance with regulatory requirements is needed (e.g., reduced emissions), or • replacement of obsolete equipment is required to meet new plant needs. The lifecycle of a project goes through many stages. The importance of each stage and its duration will vary with the project. In most cases, a project's lifecycle consists of the following processes: • define the scope and activities, • define the sequence of activities and their duration and then develop a schedule, • allocate human resources, assign roles and responsibilities, and develop an organizational chart, • estimate project cost and obtain budgets, • plan purchasing schedule to coincide with budget availability, • develop team and supply training where required, • start project and ensure proper coordination (this may involve compromises, tradeoffs, and alternatives), • complete project, and, • close project (resolve open items, project evaluation, and identify lessons learned for future projects). Quite often, once process engineering and/or researchers develop the required process, it’s tested in a lab and then implemented-first in a small-scale pilot plant and then in a full-scale plant. When it’s fully defined and the main problems are resolved, the feasibility of such a project is assessed, and if it’s successful, budgets are allocated before the work in the full-scale plant starts. Once the budgets are approved, a multi-disciplinary team is assembled to design the plant. Projects are managed by project managers who generally use proven methods when managing a project. Some of these methods are published, and some are just common sense, based on experience. A successful project manager will customize his or her project management method to the task at hand. A project manager assigns roles to the team members, allocating responsibilities and authorities and identifying the reporting relationships in a project. He or she should ensure that the group works as a team and that interaction between the different disciplines is proceeding as planned. In addition, he or she should be familiar with the technical aspects of the work to be done. A project manager should be a knowledgeable leader, a fair and dependable person, an honest and energetic individual, and a skilled negotiator. Two key limited resources control projects-budget and time. By definition, a project has a limited life span (a start and a finish) and each project is unique. Typically, in a process plant, process engineering defines the process, which becomes the basis of design. To control a project, project managers must understand the process (and project) very well. The project manager then divides the project into clear project phases or milestones. Each phase is identified with a deliverable-such as a product resulting from engineering work or construction activity. This approach creates a measurable and logical sequence in the life of a project. The scope of a project can be modified because of new government regulations or an error or omission in the original concept. These types of unforeseen delays affect budget and time and should always be allowed for. For international projects, it’s prudent to allow for extra costs and time to cover cultural and political differences, communication delays due to different time zones, different languages and the need for translation (and the possibility of misunderstandings), and the availability of on-site local technical experience. Process Control: If a comparison is made between the human body and a typical plant, the following functional similarities are found. The bones are similar to a plant's structure. The muscles are the equivalent of electrical motors (taking their command for action from the brain and the nervous system). The veins correspond to the electrical wires (carrying the energy). The organs are similar to process equipment, such as reactors. The senses are the equivalent of industrial sensors (measuring the process conditions). The brain and the body's nervous system correspond to the plant's control system (a blend of human operators and control equipment). In other words, control system personnel implement and maintain the senses, nervous system, and brain of an industrial plant. Modern control systems have been accused of eliminating jobs and creating unemployment. The reply is yes and no. Fewer people are needed to do tedious repetitive operations, but such control systems actually secure jobs by preventing plant closures through the maintenance of plant efficiency and competitiveness. In the end, and in most cases, modern control systems increase the number of jobs because of the increased market, both national and international. The implementation of process control systems depends on the corporate culture and the needs of the plant, but it typically includes • defining the project and developing a plan of action, • designing the control system, • installing the equipment, • starting the control system, and • providing all information to maintenance personnel. --1 for an overview of a typical lifecycle of a process control project, with all its activities in their proper sequence. Modern industry in today's global competition requires modern and powerful systems that pro vide precise controls that are relatively simple to implement and operate. This requires planning and project control that must meet schedule and cost budgets. Process control is involved in most industrial applications, and this requires knowledge and experience. A correct thought process is essential to correctly implement a project, and a successful project will correctly match the process requirements with the control system. The quality of the documentation produced by engineering is vital in the construction and maintenance of a facility. It allows others to pick up the project where designers left off, and it provides the reasoning as to why a decision was made (keeping in mind that hundreds or even thousands of different components form the ingredients of a control system). Unfortunately, it’s quite common that project descriptions are not sufficiently detailed, and therefore, many reviews, evaluations, and revisions typically occur. A final document is never really final. It’s vital for the success of a project to allow sufficient time to clearly agree on the scope of a project and confirm that agreement in a document. For process control, the scope of a project is confirmed in the scope definition, PIDs, and logic diagrams. Engineering activities for process control typically start at the beginning of a project with the development of PIDs, control philosophies, and logic diagrams. Also, process control typically is last to finish in the construction schedule, after most civil, mechanical, and piping work is completed. This situation creates excessive pressure on the process control construction and commissioning teams as the end approaches. This situation occurs because of a lack of funds as the project nears completion and delays generated from other disciplines. Therefore, good planning and engineering (both front-end and detailed) are vital to a successful implementation. On a typical multi-disciplinary project, process control interfaces with many disciplines. At the onset of a project, most of the interface is with process engineering for the development of all the front-engineering activities and with project management for budgeting and scheduling. Later on, when detailed engineering starts, process control interfaces with all other disciplines, such as mechanical (e.g., for connecting and mounting the equipment), electrical (e.g., for wiring and conduit runs), and even civil (e.g., for control room requirements). It’s strongly recommended that data transfer (e.g., obtaining process condition at instruments) and important communications be always done in writing. -- ___1Typical lifecycle of a process control project. 1st phase: 1. Project manager assigned, preliminary project (and budget) defined, feasibility studies completed. 2. Preliminary project approved and budgets allocated. 3. Process engineering develops material balance sheets. 4. Project manager assigns lead engineers for all disciplines. 5. Process engineering and process control engineering personnel develop preliminary PIDs and logic diagrams. 6. Lead engineers review the scope of the project and change scope where required. 7. Electrical engineering establishes electrical area classification. 8. Project manager submits to management an overall project schedule, a reviewed project definition, and a project budget (at ± 30%). 2nd phase: 9. Management approves a ± 30% budget. 10. Preliminary (front-end) engineering starts to: • prepare process control schedule, • prepare control scope definition and identify preferred vendors, • prepare preliminary instrument index, • update PIDs and logic diagrams as required, and… • review cost estimate and resubmit at ± 20%. 11. Management approves a ± 20% budget. 12. Detailed process control engineering starts to • prepare process data sheets and forwards to the process engineer to complete process information, • review mechanical and piping specifications, • prepare all detailed design documentation, • establish interface between process control engineering and electrical engineering, • supply information to electrical engineering (e.g., power supply requirements and cable runs), • supply information to mechanical engineering (to mount in-line devices), and • prepare requisitions, evaluate bids, select vendors, and place orders. 13. Check completed systems (e.g., control systems, analyzer systems, and panels and cabinets) at vendor's facility. 14. Equipment delivery 15. Construction starts instrument installation, checkout, commissioning, and loop tuning. 3rd phase: 16. Plant startup 17. Engineering is completed. 18. Control system is handed over to operations. --- Communication: Good communication is vital to the success of a project. Each project manager has his or her own style. The following points are described because they summarize key ingredients of good communication in project management and engineering. Written agreements record what was said and decided, where as verbal agreements may be for gotten, misunderstood, or modified (intentionally or not) to suit a person's interest. Verbal agreements may result in expensive corrections meaning additional money and delays. This approach to written records should apply not only to decisions made but also to the transfer of data and documents. It’s preferred that discussions occur with a recipient before a memo is sent to him/her. Receiving an unexpected memo can in some cases create unhappy and uncooperative relationships between the members of a project team. A written document typically should start with a reference subject. Such a document, in addition to indicating who the sender and recipient are, often indicates who it should be copied to, including if a copy should be sent to "file." Don’t copy persons that have no interest in the subject matter. Typically, reports have a front cover and should be reviewed by another person or by manager, especially in the case of: • new designs, • safety-related design, • items of high economic importance, or • information affecting the management of the facility. It’s common that multi-disciplinary documents require more than one review and signatures. Standard and Code Compliance: Different standards and codes are applied when implementing a project. It should be determined at the beginning of a project which standards and codes apply. Some of them are corporate (or plant), and some are from external organizations. It’s a good practice to establish a list of such standards and codes at the onset of a project. Later on, it will be a lot simpler to ensure project compliance. Design standards and codes are generally grouped under a general umbrella for a particular country; For example, ANSI in the U.S., CSA in Canada, DIN in Germany, and IEC for Europe. In North America, the main standard and code developing organizations are: ANSI = American National Standards Institute API = American Petroleum Institute ASME = American Society of Mechanical Engineers ASTM = American Society for Testing Materials CSA = Canadian Standards Association FM = Factory Mutual IEEE = Institute of Electrical and Electronic Engineers ISA = The Instrumentation, Systems, and Automation Society NEMA = National Electrical Manufacturers Association NFPA = National Fire Protection Association OSHA = Occupational Safety and Health Association UL = Underwriter's Laboratory ULC = Underwriter's Laboratory of Canada At the international level, ISO and IEC are developing standards that are gradually being adopted worldwide. This will simplify engineering and equipment production. The world is getting smaller; global trade is on the increase; and accordingly, engineering and equipment is being sourced and used worldwide. Control Strategy "Should our control strategy be reviewed?" This question has crossed the minds of many man agers, engineers, and operators. A need for a technical audit of the existing control system is the starting point ( __19). The result of a technical audit is then compared with the plant business strategy. That becomes the base from which a control strategy for the plant is developed. The plant business strategy should be expressed in terms of market and product needs. It should provide a plan for the present and the future, and it should identify the domain in which the organization operates now and into the future. The control strategy should be directly related to the plant's present and future business needs. It should be based on customer requirements, the competition, and the products being manufactured. Once a strategy is in place, a plan is created to identify the steps that must be taken to reach the goal of the organization. A balance must always be maintained between the control strategy and the plan because knowledge gained through the implementation of the control system and the plan will need to be updated. None of the parameters are static; every parameter is in continuous evolution and change-the market, customer needs, technology, competition, and government regulations. Note that once the process has started, it should move fast. Delays generate hesitation, and hesitation generates doubt and uncertainty. Eventually nothing gets done, and to restart the whole process becomes even more difficult. Going back to the original question of when to review the existing control strategy, the following telltale signs may provide an indication. Note that in most cases the control system is not the only answer to all problems. The quality of raw materials, the capabilities of the process equipment, and employee morale are but a few examples of additional key ingredients for a successful plant operation. The review of the control strategy typically occurs following telltale signs such as: • sliding market share, • unhappy customers, • inability to keep up with the competition, • recurring emission problems, • large inventories of raw materials and finished products, • inconsistent and/or poor quality, • unreliable plant trip and alarm systems, • poor or nonexistent production data, • inflexible production and long start-up time, • poor productivity, with too much staff and high wages, • errors in transferring production data to paper, • many man-hours wasted in reading data from unreliable sources, • too much time wasted in checking manually copied data, • inability to obtain immediate feedback and production knowledge, • inflexible existing production facilities, • long setup of existing facilities, • large support staffs required for the production facilities, • increasing production costs, • inability to comply with environmental regulations, which are a top priority, • budgets cuts that prevent plant investments and improvements, • constantly changing production priorities, with heavy start-up and shutdown costs to accommodate customer delivery requirements, • inability of existing production facilities to meet the required quality of service, new product introduction, and technological know-how that are key to price setting, profit ability, and business survival, • less time to respond to market demand, • inability to deliver more specialized products, better quality, better service, better delivery times, and specialized packaging with specific delivery constraints, • customers that press the plant to accept low-volume, unprofitable orders, and • the need for quick response to market demands and changes, keeping in mind equipment failure, set-up times, operating costs, and inventory costs. It should be noted that many of these telltale signs are interrelated. For example, poor quality will produce more scrap, which increases pollution, which increases the need for raw material, which increases cost and decreases profits. Some industries have these problems and do nothing about them. They don’t survive. Others take the bull by the horns and are selling success fully to an ever expanding world market. Plant Business Strategy: The plant business strategy, as was mentioned earlier, is the base of the plant control strategy. In other words, what is the control strategy really trying to achieve relative to the business side of things? The following nine points serve as a guideline (or a starting point) for developing plant business and control strategies. 1. In general, if targeted growth and profitability are to be achieved, then several major changes may be required. There is an immediate need to get control over inventories, production, and delivery performance (including finished goods, raw materials, intermediates, and packaging). 2. For management, real-time data must be available at the plant floor level, immediately accessible to plant management, and integrated into the management information system at the plant. Also, there is a need to capture the knowledge of existing (and close to retirement) personnel in an "intelligent" system. The plant must be managed from minute-to minute not from month-to-month. Management must be able to react to the present and plan for the future instead of reacting to the past. Problems must be detected and rectified before they occur. 3. From a customer point of view, the plant must be able to compare customer complaints with the production of the different shifts. In addition, customer orders must be tracked through the on-going production. 4. Plant emissions must be reduced by a stated percentage at the end of one year and by a higher percentage by the end of the following year. 5. Production must strive for zero defects, identify those processes that add value (to be enhanced) and those that add only cost (to be eliminated), and provide production information accurately when needed (no guessing, telephoning, working with old info, or manual collection of data). For all products, on-line knowledge of quality control analysis must be available on demand. 6. The plant must improve its ability to introduce new applications, produce an increasing proportion of proprietary products, and quickly and efficiently develop new products and processes. 7. The use of modern control systems should abide by the following guidelines: A. Implement first on a small scale (i.e., a relatively low-cost and acceptable learning curve) then expand plant-wide (perhaps seek government credit through R&D for pilot application). B. Start with a process that is now driving the cost up and look for a process • with as rapid a payback as possible, • that involves the fewest number of people with the least equipment, • that produces a large amount of several end products with different market values, • with a large price differential between the feed and end products, • with expensive raw materials (or expensive cleanup cost or expensive operating cost), • with high-energy consumption, and • that is hazardous and where the operators need to be kept away from the production process. C. Start in the plant automation process to get on the automation learning curve before the competition gets so far ahead that the plant can never catch up. D. The control system is required to reduce product cost and, at the same time, increase productivity and improve quality. Using production trends (good and bad), quickly identify any deviation and then correct the situation. E. The approval and implementation of a control system must be preceded by: • a scope definition (shows benefits, justification), • a description of the control system features, • an equipment list (for the potential control system), and • manpower requirements (including training needs and re-allocation). 8. Acquire an off-the-shelf system that can be easily understood by plant personnel and well supported by the vendor. 9. The management must ensure that the operators don’t get the impression that the new system will be used to "beat on" them; it’s there to help everybody do a better job. Get the operators involved in the decision making process. Implementation of a New Control System The first step is to start with a business strategy or mission. This is a management responsibility. The business strategy is used as a reference further along the process. Before further activities are carried out, management's support and commitment to improvement must be established. Without this support and commitment, activities down the line will probably be a waste of time, the concept won’t be implemented, and future efforts toward implementation will be regarded with distrust. The next step is to assemble a dedicated team that is led by a champion. The champion will probably devote all of his or her time to the project. The team's key word is "dedicated." In other words, when the need arises for action, the team must find the time to take action. The team must not be too large. It should be multidisciplinary to benefit from different fields of knowledge. For example, a representative from management, two from engineering (one from process, one from controls), one from operations, and one from maintenance. With the champion and the team in place, the justification process can start ( __19). It will consist of three parts: 1. Identification of the plant needs based on the benefits obtainable from the control system. This activity will highlight the main features of the control system to be installed. 2. Specification of a control system based on the needs previously identified. This activity is closely followed by vendor selection. Vendors are requested to bid on the potential control system. Following the receipt of all bids, a decision analysis is performed to evaluate all control systems considered and decide which meets the plant's needs most closely. 3. Totalization of all possible costs and performance of a cost justification for the project based on the benefits received and the payback achieved. The champion then draws up an implementation plan. One word of caution: Don’t automate the easiest application; look at where the benefits are needed and where the implementation is sure of being accomplished successfully. Once the budget is approved and following the decision to implement, the problem becomes to implement it in the most effective manner. Management must be involved because resources (financial and human) will have to be allocated until completion of the project. Throughout the justification/implementation process, it may be necessary to seek advice and guidance from experienced consultants. It will also be necessary to involve operators and maintenance personnel in system selection. This is good for morale and ensures support during and after implementation. Management should assure all employees that the new system guarantees business instead of taking jobs away. Employees must be kept informed of progress. In most cases, the implementation for existing plants is first done on a small scale, which gives the benefit of going through the learning curve with minimum impact. Also, it allows management to check the justification through a follow-up before jumping into a plant-wide implementation. In some cases implementation will be done on a large scale. This is especially true in new plants. Here, experience is needed because there is no room for error. It must be done right the first time! It’s important to remember the following points: A. Implementation could take as little as a few months or as much as several years; it all depends on the scope. B. Always avoid islands of automation. Communication problems can become expensive nightmares. C. Don’t automate chaos. Scheduling and Time Management: Project scheduling comes in different formats. Regardless of the format used, the purpose of scheduling is to keep control of the project by breaking down the overall project into smaller manageable activities. Complex activities should be broken down into even smaller activities. Scheduling is an on-going activity that requires updating on a regular basis. The update frequency depends on the project and its complexity. A schedule is not only used by project management but is also used by engineering to monitor deliverables and interact with other disciplines, and by purchasing to plan their work, buy material, and schedule delivery. A schedule defines different stages and milestones in a project. It’s an essential tool to manage activities. For example, a schedule for preliminary engineering activities as they apply to process control. Each stage may need to define: • its inputs, such as documents required and standards and codes applicable, • the tools and methods required to get things done, and • its outputs, such as documents produced and activities completed. ---- ___2 Simplified schedule for preliminary engineering activities showing two milestones. AUGUST JULY EVENT * = Milestone JUNE MAY APRIL Management Approves ± 30% Budget * Prepare Schedule for Process Control Prepare Control Scope Definition and Instrument Index Update PIDs + Logic Diagram. Submit Cost Estimate ±20% -- In addition, the schedule may need to define who will do the work, when will it start and finish, what milestones are to be reached, and what is required to get an activity started to reach the milestone. Typically, events and activities are drawn on the left-hand side, and time is at the bottom (or top). When well prepared, the schedule should identify the requirements for, and effective use of, available human resources and show the activities that need to be completed before some others can start. A preliminary schedule should be generated at the start of a project (having a bird's eye view of the project), including project definition, preliminary engineering, detailed engineering, construction, commissioning, and start-up. As the project evolves and its scope becomes clearly defined, the schedule is then broken down into detail activities. The detail of the schedule is in relation to the size of the project, its importance, and its complexity. Cost Estimate: Estimating the cost of a project is essential to obtaining approved budgets. The accuracy of an estimate depends on the stage of a project. At the start of a project and until preliminary engineering is completed, a ±30% is acceptable. Past experience and previous data is typically used to fill in incomplete information. On completion of preliminary engineering, a ±20% is expected-here again, a combination of past knowledge and estimates from vendors is used to reach this level of accuracy. On completion of detailed engineering, a ±10% accurate estimate can be reached. At this point, and to reach this level of accuracy, bids have been received from vendors and contractors. Typically a project goes through three basic phases. The first phase covers the start of the project, and costs start rising but generally at a low rate. Once the project is approved, the second phase starts, and costs increase significantly over time. In this phase, engineering personnel are mobilized, work starts, and equipment is purchased and installed. The third phase typically follows the completion of construction work on site, and cost diminishes until the project is closed. ___3 Project cost expenditures vs. time. Cost estimates typically include engineering and equipment costs-and quite often installation, commissioning, and startup costs. Costs are difficult to estimate for an old plant to be retrofitted. This is due to unreliable documentation that does not reflect plant conditions and to the uncertainty of existing equipment condition and capability. From a process control point of view, an estimate is frequently generated from a list. The instrument list is a good starting point. The list can be broken down as follows: • 1st column, the equipment tag number (i.e., one line per item), 1st Phase 2nd Phase 3rd Phase Cost | Time • 2nd column, the engineering costs, • 3rd column, hardware costs, • 4 th column, installation costs (including installation material), and • 5th column, total for each line. Additional lines would cover commissioning and startup on a loop-by-loop basis. Finally, a grand total to cover the total costs. Often a safety factor is added to cover unknowns (10% to 30%, depending on the project). The safety factor includes such unknowns as out-of-country projects, expertise of project personnel, and foreign languages. A budget review typically occurs as the project progresses-confirming the previously approved budgets-or when additional funds are requested (a situation often frowned upon by management). |
| Top of Page | More PM Articles | Prev: Measurement / Control Equipment--CALIBRATION | Next: Project Management for Process Control systems--part 2 | Info-Source.us |