RAMS Role in Capital Projects. Jason Engineering Manager

Transcription

1 Speaker Bio Jason Ballen*ne holds a Bachelor Degree in Mechanical Engineering and is a Cer*fied Maintenance and Reliability Professional. At ARMS Reliability, Jason manages the mul*tude of reliability projects undertaken by ARMS Reliability Engineers in North America. This includes maintenance strategy development and op*miza*on, RAMS studies and life cycle cost analysis studies for different industry sectors. Availability Workbench is a trademark of Isograph Ltd. ARMS Reliability is a global implementa@on partner and distributor of this sobware. All material displayed in this presenta@on is for private use. You may not distribute, modify, transmit, reuse, re- post, or use the content of this presenta@on for any public or commercial purposes.

2 RAMS Role in Capital Projects Jason Engineering Manager

3 RAMS for new projects in defence, aerospace and upstream oil industry has been common for past several decades. RAMS for new projects in mining, resource and energy sector is less common. Some companies with dollar projects are now specifying RAMS analysis within their major project schedules. This looks at how and why it is done

4 To the steps and your needs to take to ensure the long term success of your assets and capital investments.

5 The Reliability Role on a Project Team! The difference between the design team and the Reliability analysts is that:! The designer asks how he/she can make the system work.! The reliability analyst asks how the system might fail to perform its intended func@on over a specified period.

6 Will the design meet the over the Is there enough redundancy? Are stockpiles, surge and buffers correctly sized? What are the bovlenecks? Will my maintenance strategy deliver what the design expects? How many resources will be required and when? How many spares will I need? How do I develop a maintenance plan for new equipment? How can I validate the OEM s recommenda@ons? 6

7 Typical Phases of a Major Capital Project Concept Prefeasibility Feasibility Execu6on Opera6on

8 Project Phases Defined by IPA FEL1 FEL2 FEL3 Business Planning Scope Development Project Planning Deliverables Define the business opportunity Conceptual process design, and a preliminary scope of work, Detailed scope of work for the project Start of detailed engineering to the end of construc@on. Project start- up Milestone Cost range (typically a 25/+40 percent es@mate). More accurate cost es@mate ( 15/+25 percent) Control Es@mate 20 to 30 percent of engineering complete Complete when the project is mechanically complete Project start- up, to steady- state opera@on, Acivi6es Technical assessment, development of a milestone schedule. Selec@ng the right alterna@ve and work scope by improving project defini@on and evalua@ng (thus, elimina@ng) alterna@ves; the emphasis has moved from business evalua@on to technical evalua@on. FEL 3 takes the basic design package for the selected alterna@ve and progresses that package to a point that will enable a control es@mate to be developed. The project team is transi@oning the project to the start- up team members. Phase Complete regardless if design capacity is reached.

9 Influence Of A Major Capital Project Lifecycle Cost Major Influence Rapidly Decreasing Influence Low Influence INFLUENCE INFLUENCE FRONT END LOADING FINAL AUTHORIZATION EXPENDITURES MECHANICAL COMPLETE EXPENDITURES COMMERCIAL OPS Concept Prefeasibility Feasibility Final Design Construct Start- Up

10 ! Basic Methods Involved in RAMS Analysis for Major Capital Projects Root Cause Analysis Reliability Block Diagrams Reliability Centered Maintenance Life Cycle Cost Analysis Fault Tree Analysis Hazop Studies

11 Root Cause Analysis Concept Prefeasibility Feasibility Final Design Construct Start- Up RCA RCA RCA RCA RCA RCA Cause and Effect Analysis applicable at all stages throughout project -Lessons learned -Problem Elimination

12 Tools of the Trade: Problem Solving Problem solving is a way of thinking and needs to be taught Apollo Root Cause Analysis - Method of Approach Step 1: Define the problem by writing the: What When Where Significance Step 2: Create a cause and effect chart For each primary effect ask why Look for causes in actions and conditions Connect causes with a caused by? Support causes with evidence Step 3: Identify effective solutions Challenge the cause and offer solutions Identify the best solution, they must: Prevent recurrence Be within your control Meet your goals and objectives Step 4: Implement the best solution 12

13 Other Applicable Reliability Methods Front End Engineering Concept Prefeasibility Feasibility Final Design Construct Start- Up Outage Unit RBD@ Equip Class Equip FMECA@ Predic@on Dom Mode RCM@ Detail Maint Plan CMMS Load HAZOP Prelim HAZOP@ P&ID LCC FTA

14 Concept STAGE PHASE ELEMENT RAMS Ac6vity RAMS Objec6ve 1 Concept 1. Establish Business Objec6ves (Reliability & Availability) 2. Ensure Equipment Reliability and Maintainability Match Design Objec6ves Reliability Block Diagram. Analyse project deliverables in business case and allocate RAMS targets

15 Tools of the Trade: Reliability Block Diagrams! Reliability Block Diagrams (RBD) illustrates how sub- systems are connected from a reliability point of view! An RBD enables an understanding of how each asset or sub- system contributes to the overall reliability performance.! An RBD contains series and parallel logic

16 Series Logic The Simplest RBD Diagram: If A has an availability of 90% then the system also has 90% availability. System fails if the single component A fails because there is no open success path between the input and output.

17 Series Logic Extended Lets extend our system to 100 blocks in series where each block has an availability of 90% Availability of simple series system = (A A x A B x A E.A N ) (Where A represents Availability) A S = = or %

18 Parallel Logic A simple system with 3 components in parallel. If any of the three components failed individually the system would not fail as a path would s@ll be open between the input nodes and the output nodes. Availability of simple parallel system = 1- (Q A x Q B x Q C.Q N ) If availability of each block is 0.9 Av = 1-(0.1*0.1*0.1)=0.999

19 Increased Complexity Using Node Voting Incorporating Series and Parallel Relationships 2 required

20 Logic Example 4 tanks arranged in a circle System failed if any two adjacent tanks are out of service Develop RBD to Represent Availability 3

21 Logic

22 Define Logic Translate process diagrams & PFD s to Reliability Block Diagram Logic Area by area basis. Redundancies. Cri6cal points of failure. Storage/surge buffers Success path logic

23 Area Area by area based on maximum throughput capacity of each area/system.

24 Storage Buffers & Surge Area by area basis. Time to Empty. Time to Fill.

25 Major Outages Unit Outage Type INT Dur Comment Bauxite Feed Major Maintenance 3 yrs 10d Single Feed String No Redundancy Grinding Circuit Mill Overhaul 2 yrs 20d Split system, full redundancy Digestion Vessel Descale 1 yr 60d Multi Trains x 3, 50% ea Boiler Boiler Overhaul 4 yrs 30d 2 units, 60% ea

26 Plant Availability/Capacity Profile Provides visibility into design capability. Drill down and identify system contributions.

27 Reliability to Design Concept Assess Project concept, availability and capacity levels Test robustness of the proposed design Single unit or unit design Ensures early visibility of reliability and maintainability issues Aids of project concepts with project and cost - 25% to +40%.

28 Ensure Equipment Reliability And Maintainability Match Design Objec6ves STEP PHASE ELEMENT RAMS Ac6vity RAMS Objec6ve Establish/Validate Equipment Hierarchy RCM/FMECA Maintainable items, required and equipment 2 Pre- Feasibility Cri6cality, Reliability and Redundancy Assessment RBD Preliminary es@mates at equipment level, planned maintenance windows, and assess design alterna@ves size/parallel units/ technology.

29 Ensure Equipment Reliability And Maintainability Match Design Objec6ves STEP PHASE ELEMENT RAMS Ac6vity RAMS Objec6ve Maintenance Define maintenance Requirements RCM requirements, and Analysis cost forecast. 3 Feasibility Value Engineering, reliability Es6mates and trade off studies. level System performance predic@ons, availability, bovleneck and redundancy studies Preliminary Hazard and Risk Analysis HAZOP Iden@i@fy Hazards, mi@ga@on and protec@on systems.

30 Level Process flow diagrams to model Failure & maintenance data gathered from best source. Build Detail of RBD model Run Evaluate and test What If scenarios

31 Process Flow Diagrams CAP DIGESTION, AREA 240 System logic paths Capacity Standby or surge arrangements FROM HID FROM GRINDING 14 FROM EVAP BAUXITE 12.5 SLURRY 24 DIGESTION 22 STORAGE LIQUOR TANKS 8 hr 2 x 1 hr VS VS MILL LIQUOR 5 SLURRY DILUTION SLURRY MIX 7.5 MIX TANK TANK 5 min 5 min TO CLARIFICATION AREA 230 VARIABLE SPEED VS VS VS POSITIVE PD PD PD B/O TO EVAP B/O TO EVAP PD PD PD DISPLACEMENT P P P P P P PUMPS R/T R/T PROCESS COND COND CC CC CC CC CC CC COOLERS (TYP LCV) LP STEAM 1 MOL MOL 1 LP STEAM VS MP MP MP MP MP MP VS A_DIG.DRW COND BACK TO BOILER HP LS HT H T HP LS HT H T HP LS HT HP H STM T HP LS HT HP STM H T HP LS HT H T HP LS HT COND BACK TO BOILER H T HOLDING TUBE (DIGESTER)

32 Major System Logic Defined RBD Systems

33 RBD Drill Down Sub- Systems

34 Further Sub Systems

35 RBD Drill Down Assets

36 RBD Drill Down Components

37 Assign Block Data ID Standby type Failure data model Item capacity age for offseung PM s Hold for repair or opportunity maintenance dependant blocks

38 Failure & Maintenance Data Failure & Maintenance Data Industry sources Experts plants manufacturers Equipment failure rates Planned maintenance requirements task requirements Equipment Review Sheet Equipment Description: Area: Equipment Block: No. Equipment items in block: 1 Capacity of each item: 55% Equipment Function: Surge Capacity Startup and Run down: Hot standby Cold Standby Maintenance Strategy PM Inspection Run to failure Information Sources Condensate pump long shaft can pump Digestion Unit / condensate system Pump LP steam condensate n/a n/a PM at annual unit shutdown Vibration test bearings every 3 months comalco Assumptions: Modeled on 1 pump No solids entrained in condensate Mtce assumed since not covered by Comalco schedule

39 Assign Failure Data type Weibull Log- normal Normal Fixed Buffer (surge)

40 Define Task type to repair Standard Miscellaneous costs Age factor Spares & Labour

41 Define Planned Task type Mean Standard Miscellaneous costs Age factor Spares & Labour Opportunity maintenance minimum age

42 Define Task Out of service during type Mean Standard Miscellaneous costs Age factor Labour interval P- F (warning) interval

43 Assign Consequence Assign to system and sub- systems in the model

44 System Capacity All shutdowns, failures and planned maintenance Overall mean capacity

45 Sub System Capacity All shutdowns, failures and planned maintenance Sub System mean capacity

46 Unit Capacity tasks Tasks Unit mean capacity

47 Improvement Sort by capacity Review low capacity blocks in model ID Description Me a n Ca pa city Digestion Digestion System Tank Tank Pump Pump Piping Piping and Valves Filters Filters Largest Impact on Capacity

48 Evaluate Improvement Install tank bypass Add a spare pump Evaluate alternate supplier

49 Improvement Note no system outages due to tank or pump

50 Data Quality Generic Performance Characteristics RBD (Availability Capacity Modeling) Data Used for Predictions Specific Characteristics RCM (Maintenance Strategy Development) Generic Library RCM Data Project Life

51 Available RBD Data Concept Phase High level system outage data Pre Feasibility Phase Unit outage data and major maintenance windows Feasibility Equipment class data and dominant failure mode. Detailed Design Equipment specific data, vendor data or other user data. Actual failure data

52 RBD Summary Reliability Block Diagrams are used in System analysis to: simplify complex systems, provide a means to predict system reliability performance and, aid in evalua@on of alterna@ve configura@ons.

53 Tools of the Trade: HAZOP Hazard and Operability Study HAZOP is a fundamental hazard iden@fica@on technique which systema@cally evaluates each part of a system to see how devia@ons from the design intent can occur and whether they can cause problems.

54 HAZOP Produce a full descrip@on of the facility/process and the intended design condi@ons. Review every part of the facility/process to discover how devia@ons from design inten@on can occur. Decide whether these devia@ons can lead to hazards or operability problems. AS/NZS3931:1998 Risk Analysis of Technological Systems Applica@on Guide 54

55 HAZOP Steps 1. Define and scope of the study. 2. Assemble the team. 3. Collect the required drawings and process 4. Analyse each major item of equipment and all equipment using assembled team and collected 5. Document likelihood and consequences of any from normal, and document those considered hazardous and credible. means to detect and prevent the AS/NZS3931:1998 Risk Analysis of Technological Systems Guide 55

56 HAZOP " Risk ESTABLISH THE CONTEXT The strategic context The organisational context The risk management context Decide the structure Develop Criteria IDENTIFY RISKS What can happen? How can it happen? ANALYSE RISKS Determine existing controls Determine likelihood Establish level of risk ASSESS RISK Determine consequences MONITOR AND REVIEW Compare against criteria Set risk priorities TREAT RISK Identify treatment options Evaluate treatment options Prepare treatment plans Implement plan 56

57 HAZOP Summary method of hazards and operability problems across an facility. Useful in unforeseen hazards due to lack of Useful in hazards introduced by changes to such as plant process or procedures. Can be applied to plants or to assist in safer detailed design during design stages. Originally developed for the chemical industry by DuPont, but has been successfully adapted in a range of industries. 57

58 Ensure Maintenance Strategies Match Equipment Reliability And Maintainability Requirements STEP PHASE ELEMENT RAMS Ac6vity RAMS Objec6ve 4 Execu6on: Detailed Engineering Maintenance Requirements Defini@on. RCM analysis Task and resource Es@mates for correc@ve, preven@ve, inspec@on and shutdown work documented. Op@mised maintenance task selec@on and maintenance plan development. Work Instruc@ons prepared

59 Prob of fail Tools of the Trade: RCM Reliability Centered Maintenance is a structured process that drills down to Failure Modes and provides the basis for choosing maintenance tasks. Traditional view of Maintenance Factor of Safety Prob of fail What if shape of curve is infant mortality? Infant mortality Age Age Wear out Maintain Prob of fail Increased risk at increased cost! Maintain Age Birthplace of RCM was to know the failure curve of each failure mode before deciding on which task was applicable. Then decide if it is worth doing. 59

60 Predict the Behavior Probability of Failure Infant Mortality Random Failure Wear Out Beta < 1 Beta = 1 Beta >1 Inspection to identify Potential failure or RTF Avoid Preventive tasks Investigate design Review parts & installation Develop commissioning plan Time Preventative or Predictive task

61 Intro to RCM RCM # Systema@c approach # Focuses on preserving func@on # Addresses failures that maver # Acknowledges Design Limita@ons # Driven by Safety and Economics # Defines Failure as Any Unsa@sfactory Condi@on # Uses a Logic Tree to Screen Maintenance # Tasks Must Be Applicable and Worth Doing 61

62 Seven of RCM 1. What are the Func6ons of the asset? What purpose does the equipment serve? 2. What Func6onal Failures are likely to occur? What problems can occur? 3. What causes each func6onal failure? What failure modes cause the problems? 4. What happens when each failure occurs? What are the effects of the failure? Safe, Env, Op, $ 5. In what way does each failure ma`er? How significant are the Effects? Severity x Rate 6. What should be done to predict or prevent each failure? What predictive or preventive maintenance can be done? 7. What should be done if a suitable proac6ve task cannot be found? What s the default action? 62

63 FMEA Analysis FMEA Drill down. # Maintainable Item # Functions # Functional Failures # Failure Modes are the Physical Causes of Failure. 63

64 RCM Decision Logic Is the failure evident in the normal course of operations? no yes Will the failure have a direct and adverse effect on environment, health, safety? yes no yes Will the failure have a significant cost impact. Is there an effective Inspection or Monitoring task? no Is there an effective Preventive task? no no yes yes Develop Inspection Task Schedule and/or monitoring scheme. Develop Preventive Task and Schedule. Redesign or Accept Risk & Run to Fail Run To Fail 64

65 RCMCost Assign Failure Effects Associate Weibull curve Estimate Repair Resources and Duration Choose Best Task Do Nothing Preventive Task Predictive Task Redesign 65

66 Outputs: Task Frequency Advice for decision making based on performance simulation. Redundancy modeling. Cost Benefit comparison for alternative maintenance strategies. Low Maint Costs Hi Failure Costs Make your decisions based on measurable outcomes. 66

67 Best Task- Worth Doing? Fix it? # Decide op@mum maintenance tasks # Determine spare part requirements # Determine maintenance schedules. # Predict resource requirements Preventive Task? Inspection Task? 67

68 RCM Delivers A safe opera6on Ensures plant reliability & efficiency Provides a document base for planned maintenance Predicts resource requirement Predicts spares usage Predicts maintenance budget

69 The Work Plan 69

70 Forecasted Costs Effects ID Description Cost Per Hour Cost Per Occurrence Safety Severity Occurances Per year Cost Per Year 10 Year Cost Safety Criticality unav Blower unavailable $269 $2,690 0 red Reduced Capacity $723 $7,226 0 saf Noise $0 $ Labour ID Cost Per Hour Corrective Call-Out Cost Hours per year Lifetime Cost MECH $16,484 ELEC $9,246 Spares ID Unit Cost Number Used Over Lifetime Lifetime Cost MOTOR $6,854 IMPELLER $7,962 BEARING $1,101 BELT $377 CASING $1,352 70

71 Maintenance Budget $900,000 Maintenance Budget $800,000 $700,000 $600,000 Cost $500,000 $400,000 $300,000 $200,000 $100,000 $ Total_inspection $11,777 $12,129 $12,957 $11,948 $13,047 $12,129 $12,189 $12,761 $12,196 $12,988 Total_preventive $240,635 $301,216 $321,091 $331,020 $338,157 $341,106 $338,578 $352,517 $347,308 $354,271 Total_corrective $344,267 $383,381 $407,990 $430,074 $444,113 $440,354 $441,378 $443,461 $440,782 $453,329 71

72 RAMS Analysis STEP PHASE ELEMENT RAMS Ac6vity RAMS Objec6ve Hazard HAZOP analysis. Detailed hazard and operability review 4 Execu6on: Detailed Engineering Quan@ta@ve Risk analysis, SIL determina@on and Protec@ve Systems design. FTA Analysis Iden@fy important risk items and risk mi@ga@on requirements. Analysis of vendor submissions Life Cycle Cost Compare Vendor submissions for reliability, avaioability, maintainability and total lifecycle costs

73 Tools of the Trade: Fault Trees Developed by Bell in 1962 to evaluate the ICBM launch control system Graphically represent interaction of failures TOP events Basic events Linked via logical symbols (gates)

74 Failure vs Success Logic Normal to construct using failure rather than success, as it is the failures that are of interest A system usually contains less fault trees that success trees Easier to analyse a fault tree rather than a success (dual) tree, as probabili@es tend to be lower Some trees contain a mixture of failure and success states

75 Top Events! Total loss of production! Safety system unavailable! Explosion! Loss of mission! Toxic emission

76 Basic Events! Pump failure! Temperature controller failure! Switch fails closed! Operator does not respond

77 Gates

78 Example

79

80 Fault Tree Summary

81 Tools of the Trade: Life Cycle Cost Analysis Acquisition Costs Sustaining Costs Disposal Costs R&D Initial Investment Recurring Investment Recurring Investment Recurring Investment Operational costs Maintenance & Reliability Costs Disposal

82 Life Cycle Cost Analysis Purchase Cost vs. Life Cycle Cost: $ Lowest Cost of Ownership Acquisition Costs Sustaining Costs Disposal Costs Least Purchase Cost

83 Cost Comparison

84 Usefulness of Life Cycle Analysis select the best out of several proposed value of money when considering capital versus ongoing maintenance. Event Timings When to replace / overhaul assets Sensi@vity Boundaries How do the costs / benefits change with changes to assump@ons 84

85 RAMS in Capital Projects Reliability Engineering = Life-cycle Management! Conceptual Define functional requirements Set reliability, maintainability, availability, safety (RAMS) targets Use like-type data from similar plant, databases, experts to test design Detailed Design Develop maintenance strategy, spares, manning Confirm reliability of detailed design Estimate life costs and compare cost/benefit of options Evaluate SH&E risk levels Develop contractor warranties Execution 95% of life costs are determined by engineering decisions made before plant is built! Build plant Implement maintenance strategy Develop performance metrics and management processes Operation Audit performance, solve problems Refurbishment/upgrades, Continuously improve reliability, reduce cost and SH&E risk

86 Reliability Engineering A RAMS Analysis provide a common basis for intelligent decisions. Simple decision making Easily updated Captures current knowledge Data driven Encourages innova@on what if Performance oriented Business decisions drill down to the lowest level. 86

87 What does it mean to Asset Managers? Know what the maintenance plan is doing for them. Can forecast their budget requirements Demonstrate compliance to risk and cost management. Forecast and predict asset performance. Plan for improvement 87

88 What does it mean to Asset Owners? Confidence in asset management plans. Risk Management Focus on areas. Eliminate non value adding areas 88

89 Your What are the barriers? Which elements of your should/ could perform RAMS analysis? 89

90 Questio ns? JASON BALLENTINE Engineering Manager