3.4 Application Modelling of Reliability and Performance in Engineering Design 243 In any complex process plant, there are literally thousands of different systems, sub-systems, assemblies and components, which are all subject to failure and, there- fore, require specific attention with respect to the integrity of their design, design configuration as well as integration. To determine a logical starting point for any RAMS analysis, a hierarchical approach is first adopted, followed by identification of those items that are considered to be cost or process critical. Cost critical items are the relatively few systems items of which the engineer- ing costs (development, operational, maintenance and logistical support) make up a significant portion of the total costs of the engin eered installation. Process critical items are those systems items that are the primary contributors to the con tinuation of the mainstream production process. Determination of cost and process criticality should begin at the higher hierar- chical levels of a systems breakdown structure (SBS), such as the plant/facility level, since the total plant is normally broken down into logical operations/areas relating to the production process. Thus, rather than simply starting a RAMS analysis at one end of the plant and progressing through to the other end, focus is concentrated on specific areas based on their cost and process criticality. The Pareto principle is followed, which implies that 20% of the plant’s areas contribute to 80% of the total engineering cost. When determiningprocess criticality,the fundamentalmainstream processes should first be identified based o n the process flow and status changes of the process. All operations/areas in which the process significantly changes, and which are critical to the overall process flow, must be includ e d. The different criti- cal processes are then compared to those ope rations/areas identified as cost critical, to identify the sections or buildings (in the case of facilities) that are process critical but may not be considered as cost critical. With such an approach, the RAMS analysis can proceed in a top-down progres- sive clarification of the plant’s systems and equipment, already with an understand- ing of which items will have the highest criticality in terms of cost and process losses due to possible failure. As a result, the RAMS analysis deliverables can be summarised as follows: RAMS activities Deliverables First-round costing Estimate initial maintenance costs Process definition Develop operating procedures Develop plant shutdown and start-up procedures Pre-commission Equipment register Initial equipment lists Equipment technical specifications Manufacturer/supplier data Plant defin ition Equipment sy stem s hierarchy structures Equipment inventory and systems coding Consolidated equipment technical specifications and group coding FMEA Failure modes, causes and effects matrices Failure diagnostics trouble-shooting charts 244 3 Reliability and Performance in Engineering Design RAMS activities Deliverables Identification of certified and critical equipment (FMECA) Critical equipment lists Plant safety requirements Process reliability evaluation Risk management directives Spares requirements planning (SRP) BOM and catalogue numbering Spares lists and critical spares Suppliers, supply lead times and supply costs Maintenance standard work instructions (SWI) Relevant statutory requirements Safe work practices Required safety gear Design updates and/or reviews Equipment modification review Interdisciplinary participation Plant procedures Statutory safety p rocedures Maintenance procedures Maintenance tasks per discipline/equipment Maintenance procedures sheets and coding for work orders cross referencing Plant shutdown procedures Plant shutdown tasks per discipline and per equipment Manning requirements Maintenance task times Maintenance trade crew requirements Maintenance budgeting Manning/spares costs against estimated maintenance tasks The RAMS analysis application model is object-oriented client/server database tech- nology initially developed in Microsoft’s Visual Basic and Access. The model con- sists of a front-end user interface structured in OOP with drill-down data input and/or access to a normalised hierarchical database. The database consists of several keyword-linked data tables relating to major development tasks of the RAMS anal- ysis, such as equipment, process, systems, functions, conditions tasks, procedures, costs, criticality, strategy, SWI (instructions) and logistics. These data tables relate to specific analysis tasks of the RAMS model. The keywords linking each data ta- ble reflect a structured six-tier systems breakdown structure (SBS), starting at the highest systems level of plant/facility, down to the lowest systems level of com- ponent/item. The SBS data table keywords are: plant, operation, section, system, assembly, component. Database analysis tools, and database structuring in an SBS, enables the user to review visual data references to specific record dynasets in each of the data tables, as illustrated in Fig. 3.51. Database structuring in an SBS, and the normalising of each dynaset of hier- archical structured records with a unique identifier (EQUIPID), allows for the es- tablishment of a normalised hierarchical database. These dynasets include specific analysis activities such as: 3.4 Application Modelling of Reliability and Performance in Engineering Design 245 Fig. 3.51 Database structuring of SBS into dynasets • PFD (process flow diagrams), • P&ID (pipe and instrument diagrams), • technical specifications, • process specifications, • operating specifications, • function specifications, • failure characteristics/conditions, • fault diagnostics, • equipment criticality and performance measures, • operating procedures, • maintenance procedures, • process cost m odels, • operating/maintenance strategies, • safety inspection strategies, • standard work instructions, • spares requirements. In designing hierarchical relational database tables, database normalisation min- imises duplication of information and, in so doing, safeguards the database against certain types of logical or structural problems, specifically data anomalies. For 246 3 Reliability and Performance in Engineering Design example, when multiple instances of information pertainin g to a single item of equipment in a dynaset of hierarchical structured records occur in a data table, the possibility exists that these instances will not be kept consistent when the data within the table are updated, leading to a loss of data integrity. A table that is sufficiently normalised is less vulnerable to problems of this kind, because its structure reflects the basic assumptions forwhen multiple instances of thesame informationshouldbe represented by a single instance only. Higher degrees of normalisation involve more tables and create the need for a larger number of joins or unique identifiers (such as EQUIPID), which reduces performance. Accordingly, more highly normalised tables are used in database applications involving many tran sactions (typically of the dynasets of analysis activities listed above), while less normalised tables tend to be used in database applications that do not need to map complex relationships between data entities and data attributes. The initial systems hierarchical structure, or systems breakdown structure (SBS), illustrated in the RAMS analysis model in Fig . 3.52 is an overview location listing of the plant into the following systems hierarchy: Systems hierarchy Description Plant/facility Environmental plant Operation/area Effluent treatment Section/building Effluent neutralisation The initial systems structure of an engineered installation must inevitably begin at the higher hierarchical levels of the systems breakdown structure, which constitutes a ‘top-down’ approach. However, such an SBS will have already been developed at the engineering design stage and, consequently, a ‘bottom-up’ approach can also be considered, especially for plant analysis of components and their failure effects on assemblies and systems. The initial front-end structurin g of the plant begins with the identification of operation/area, and section/building groups in a systems breakdown structure. As illustrated in Fig. 3.53, this structuring further provides visibility of process sys- tems and their constituent assemblies and components in the RAMS analysis model spreadsheets, process flows and treeviews. Relevant information can be hierarchi- cally viewed from system level, down to sub-system, assembly, sub-assembly and component levels. The various levels of the systems breakdown structure are nor- mally determined by a framework of criteria that is established to logically group similar components into sub-assemblies or assemblies, which are then logically grouped into sub-systems or systems. This logical grouping of the constituent items of each level of an SBS is done by identifying the actual physical design configu- ration of the various items of one level of the SBS into items of a higher level of systems hierarchy, and by defining common operational and physical functions of the items at each level. The systems hierarchical structure or systems breakdown structure (SBS) is a complete equipmen t listing of the plant into the following hierarchy with related example d escriptions: 3.4 Application Modelling of Reliability and Performance in Engineering Design 247 Fig. 3.52 Initial structuring of plant/operation/section Systems h ierarchy Description Plant/facility Environmental plant Operation/area Effluent treatment Section/building Effluent neutralisation System/process Evaporator feed tank Assembly/unit Feed pump no.1 Component/item Motor–feed pump no.1 Figure 3.54 illustrates a global grid list (or spreadsheet) of a specific system’s SBS in establishing a complete equipment listing of that system. The purpose for describing the systems in more detail is to ensure a common understanding of exactly where the boundaries of the system are, and which are the major sub-systems, assemblies and components encompassed by the system. The boundaries to other systems and the interface components that form these bound- aries must also be clearly specified. This is usually done according to the most ap- propriate of the following criteria that are then described for the system: • Systems boundary according to major function. • Systems boundary according to material flow. 248 3 Reliability and Performance in Engineering Design Fig. 3.53 Front-end selection of plant/operation/section: RAMS analysis model spreadsheet, pro- cess flow, and treeview • Systems boundary according to process flow. • Systems boundary according to mechanical action. • Systems boundary according to state changes. • Systems boundary according to input, throughput or output. Interconnecting components such as cabling and piping between the boundaries of two systems should be regarded as part of the system from which the process flow emanates and enters the other system’s boundary. The interface components, which are those components on the systems boundary, also need to be clearly specified since it is these components that frequently experience functional failures. Also, systems such as a hydraulic system, for instance, may not contain all the compo- nents that operate hydraulically. For example, a hydraulic lube oil pump should rather be placed under the lubrication sub-system. Where each assembly or a com- ponent is placed in the SBS should be based on the criteria selected for boundary determination. Normally for process plant, the criteria would typically be that of inputs and outputs, so that the outputs of each assembly and component contribute directly to the outputs of the system. 3.4 Application Modelling of Reliability and Performance in Engineering Design 249 Fig. 3.54 Global grid list (spreadsheet) of systems breakdown structuring The selected system is then described using the following steps: • Determine the relevant process flow and inputs and outputs, and develop a pro- cess flow block diagram, specifically for process plant. • List the major sub-systems and assemblies in the system, b ased on the appropri- ate criteria that will also be used for boundary determination. • Identify the boundaries to other systems and specify the boundary interface com- ponents. • Write an overview narrative that briefly describes the contents, criteria and boundaries of the systems under description. A complete equipment listing of a plant includes the following activities at each systems hierarchical level: Equipment listing at system level provides the ability to: • identify g roups of maintenance tasks for m aintenance procedures, • identify g roups of maintenance tasks for maintenance budgets, • identify critical systems for plant criticality, • identify critical systems for maintenance priorities, • identify critical systems for plant shutdown strategies. 250 3 Reliability and Performance in Engineering Design Equipment listing at assembly level provides the ability to: • identify location of pipelines, • identify location of pumps, • give codes to pumps, lube assemblies, etc., • identify critical assemblies for maintenanc e strategies. Equipment listing at component level provides the ability to: • identify relevant technical data of common equipment groups, • identify relevant technical data to establish bill of materials groups, • identify and link bill of spares, • identify critical components for spares purchase, • identify location of in strumentation, • identify location of valves, • give codes to classified/critical manual valves, • identify required maintenance tasks, • establish necessary standard work instructions, • establish necessary safe work practices, • give codes to valves for operation safety procedures, • give codes to MCC panels, gearboxes, etc. A process flow diagram (PFD), as the name implies, graphically depicts the process flow and can be used to show the conversion of inputs into outputs, which subse- quently form inputs into the next system. A process flow diagram essentially depicts the relationship of the differentsystems and sub-systems to each other, based on ma- terial or status changes that can be determined by studying the conversion of inputs to outputs at the different levels in each of the systems and sub-systems. One reason for drawing process flow diagrams is to determine the nature of the process flow in order to be able to logically determine systems relationships and the different hierarchical levels within the systems. Most process engineering schematic designs start off with simple process flow diagrams, as that illustrated in Fig. 3 .55, from which material flow an d state changes in the process can then be identified. This is done by studying the changes from inputs to outputs of the different systems and determining the systems’ boundaries as well as the interface componentson these boundaries. A side benefit is a complete description of the system. The treeview option enables users to view selected components in their cascaded systems hierarchical treeview structure, relating the equipment and their codes to the following systems hierarchy structure: • parts, • components, • assemblies, • systems, • sections, • operations, 3.4 Application Modelling of Reliability and Performance in Engineering Design 251 Fig. 3.55 Graphics of selected section PFD • plant, • site. Figure 3.5 6 illustrates a typical treeview in the RAMS plant analysis model with expanded SBS (cascaded systems structure) for each system. The RAMS analysis list is a sequential options list of the major development ac- tivities and specifically detailed specifications of a system selected from the section process flow diagram (PFD). By clicking on the PFD, a selection box appears for analysis. The options listed in the selection box in Fig. 3.57 include the following analysis activities: • Overview • Analysis • Specifications • Diagnostics • Modifications • Simulation • Decision logic • Planning • SWIs • Procedures • BOMs • Technical data • Grid list • PIDs • Reports • Treeviews 252 3 Reliability and Performance in Engineering Design Fig. 3.56 Graphics of selected section treeview (cascaded systems structure) The first categoryin the RAMS analysis list is an overview ofspecifically detailed technical specificationsrelating tothe equipment’sSBS, specifications,function and requirements, including the following: • Equipment specifications • Systems specifications • Process specifications • Function specifications • Detailed tasks • Detailed pro cedures • Logistic requirements • Standard work instructions. Figure 3.58 illustrates the use of the overviewoption and equipment specification information displayed in the equipment tab, such as equipment description, equip- ment nu mber, equipment reference and the related position in the SBS data table. The technical data worksheet illu strated in Fig. 3.59 is established for each item of equipment that is considered during the design process to determine and/or mod- ify specific equipment technical criteria such as: . Reliability and Performance in Engineering Design example, when multiple instances of information pertainin g to a single item of equipment in a dynaset of hierarchical structured records occur in a. database tech- nology initially developed in Microsoft’s Visual Basic and Access. The model con- sists of a front-end user interface structured in OOP with drill-down data input and/ or access. requirements. In designing hierarchical relational database tables, database normalisation min- imises duplication of information and, in so doing, safeguards the database against certain types of logical