Reliability Engineering in Biomanufacturing - Presentation by Michael Andrews
- 1. Reliability Engineering in Biomanufacturing Michael K. Andrews Senior Reliability Engineer Bristol-Myers Squibb Manufacturing, Clinical, R&D
- 2. Overview • Definition • Purpose • Emergence of Reliability Engineering • The need • Skills Where does it all go wrong? • Challenges • Obstacles • Impacting Factors • Industry Data Reality • Reliability in Design Phase Applying Reliability Principles • Applying Reliability Principles • Tools and Methods: RAM, FMEA, RCA Agenda:
- 3. Defining Reliability “System or component to function as intended, under stated conditions, for a specified period of time” • We expect our car to start every morning • We expect the plane to arrive safely • We expect food to be on shelves in grocery store • We expect clean drinkable water from the faucet
- 4. To identify, manage and mitigate asset/process risks that could adversely affect plant or business operations Purpose of Reliability Engineering Main focus areas: • Risk Management • Loss Elimination • Life Cycle Asset Management (LCAM)
- 5. Emergence of Reliability Engineering 1940-1950 - Fix it when it breaks 1960-1970 - Higher plant availability - Longer Equipment Life - Lower Costs 1980-2000 - Higher plant availability and reliability - Greater Safety - Better product quality - No environmental impact - Longer equipment life - Greater cost effectiveness Prior to WWI industry was not highly mechanized, high availability of manpower. Manufacturing of equipment was simple and lower skills needed to maintain. WWII manpower became limited. Manufacturing became more mechanized and complex. Dawn of preventative maintenance and soring maintenance cost and expensive capital machines. Highly mechatronic systems (computerized integrated with mechanical) and extremely complex system interdependencies. Niche skills develop and a loss of overall cohesion and mission focus deteriorate.
- 6. • Manufacturing costs continue to rise. Managing Risk is essential to controlling costs. • Expanding Regulations (environmental, health, safety and quality) impact cost. Preventing/reducing incidents are critical to avoid fines and branding damage. • Reliability has evolved beyond a maintenance function. It is essential to have it tied to the business strategy which will affect Total Cost of Ownership. • Shortage of Reliability Engineers who require knowledge of multiple engineering disciplines (Mechanical, Process, IT, Automation and Business) • Overall shortage of engineers and technical trade skills The Need
- 7. • Develop a reliability and maintainability (R&M) plan and goals, evaluate field data for R&M improvement • Data analysis, trending and other data mining to find solutions to chronic problems • Manage asset risk as related to health, safety and environment (HES), production, quality, regulatory compliance and cost. • Be an integral part of the design, installation and buy off of assets - for new assets and major changes to assets. This is important because over 90 percent of the lifecycle cost is decided early in a project. • Take systems-thinking approach (machinery, equipment, controls, processes, utilities, safety, environmental, people and more) to ensure reliability • Monitor production and maintenance losses to improve throughput and reduce cost (MTTR, MTBF and Availability) • Conduct failure mode and effects analysis (FMEA) • Perform maintainability prediction calculation and demonstration • Develop and implement R&M specifications for purchasing, apply to meet R&M design requirements. • Perform reliability modeling to make trade-off decisions (what to implement) • Use R&M analysis tools to identify low reliability and high level of maintenance components/tasks to enable redesign of selecting other approaches. • Uses historical maintenance data, production data, safety data and other sources to deliver a comprehensive strategy for improvements. • Protects people from unexpected outcomes. • Evaluates a situation and knows whether or not to use maintenance predictive technologies, condition based monitoring, time based intervention, or run to failure. Skills Source ReliabilityWeb.com
- 8. Many companies struggle and unfortunately fail at implementing effective Reliability strategies due to: Challenges in Applying Reliability Principles 1) Lack of Culture, Vision, connecting Business Plan 2) Dedicating resources needed 3) “Short Term” memory 4) Inability to let go of myths
- 9. • Culture, Vision, Plan – Reliability is holistic approach requiring “buy-in” from all departments. “Reliability” is everyone's job. Leaders must understand – believe – reinforce! This must be tied to Business Strategy. • Resources – Silo mentality of companies – departments become mini cost centers unwilling to dedicate resources. Cross functionality input is required to be successful. • “Short Term” memory – Management inability to stick with long term strategies due to fiscal year blinders and only focus on “cost cutting”. This is a fundamental lack of understanding the Impacting Factors. • Inability to let go of myths – Companies tend to resist change. This is coupled with a risk adversity mainly because they don’t truly understand their risk position and how it impacts Total Cost of Ownership. Biomanufacturing tends to be PM heavy and Quality departments can be overly conservative making it a greater challenge. Obstacles to overcome!
- 10. • Loss of market share • Unit cost too high • Maintenance cost too high • Chronic product quality issues • Excessive downtime • Variability in processes and practices • Inadequate budgets • Not enough people • Undependable suppliers Impacting Factors A comprehensive Reliability Program aims to address these factors
- 11. Impacting Factors • Cost Avoidance – Elimination of unnecessary costs, such as overtime, contract labor, etc. – Elimination of losses and waste, such as scrap, excess energy usage, etc. • Increased Revenue – Improved Asset Utilization (AU) or OEE – Elimination of unnecessary planned and unplanned downtime
- 12. “A” Companies that implore some Reliability Principles effectively manage Unit Cost. “B” Companies struggle and can be close to failing - it may only take certain market conditions to drive them under if they don’t change. “C” Companies have passed the inflection point and exceeded the ability for turnaround and are on path to going out of business. Data reality: Business Extinction Source LifeCycle Engineering
- 13. Data reality: Causes of un-reliability Companies still to hold on to the myth and focus on “costs” due to maintenance. Belief that equipment breakdowns are due to this. They focus on cost managing this area largely ignoring the other 83% causes. Source LifeCycle Engineering
- 14. Data reality: Causes of lack of production Utilization Companies also hold on to the myth belief that lack of production utilization is due to poor maintenance, largely ignoring the other 93% of contributing factors. Source LifeCycle Engineering
- 15. 80% Design Decisions Impact Total Life Cycle Cost. Proper Design for Reliability applied in first design phase can account for 40% reduction in Capital Cost Unreliable Design: Capital Cost Souce: © 2011 WERF, Water Research Foundation, GWRC, GHD Consulting Inc.
- 16. 95% of Life Cycle Cost is determined before the equipment first turns on Unreliable Design: Life Cycle Cost 95% Source: Engineering Economic Analysis
- 17. Changes to fix design issues later become difficult and very expensive! Reliable Design: Changing after fact Source: Engineering Economic Analysis
- 18. Applying Reliability Process RAM Modeling User Requirements Engineering Design Specs Pre-Design Final Design Operation Criticality Assignment / Maintenance Strategy Maintenance Implement Maintenance Strategy / Design Change KPI’s & Metrics Periodic Alignment Business Strategy Failure Trending Failure Event RCA Change Maintenance Strategy / Re- Design? Build/Install Commissioning & Validation YES
- 19. RAM (Reliability, Availability and Maintainability) RAM modeling is paramount in plant and process design. When done in the design stage Total Cost of Ownership can be understood. This is done by modeling process flow and throughput, availability, critical failure points, maintenance strategy and costs. Source: SoHar, inc
- 20. FMEA (Failure Modes and Effect Analysis) Source: Megha Thakkar, NDDS, Cipla, LTD Risk Priority Number = Severity x Occurrence x Detectability The FMEA starts with defining the Risk Matrix which serves as the criteria for assessing the Functional Failures A RPN (Risk Priority Number) is calculated by taking into account the three variables: Severity, Detectability and Likelihood of Occurrence
- 21. FMEA Source: Mark A. Morris, ASQ Automotive Division
- 22. FMEA Example Source: Mark A. Morris, ASQ Automotive Division
- 23. Root Causes Analysis Method / Tool? Defines Problem? Defines all known Causes? Provides a Casual Path to Root Causes? Delineates Evidence? Explains How Solutions Prevent Recurrence? Score Apollo Method Method Y Y Y Y Y 6 Fault Tree Method Y Y Y N Y 4 FMEA Tool Y N Limited N Limited 2 5 Why’s Method Y N Y N N 2 Events & Casual Factors Method Y Limited N N N 1.5 Pareto Tool Y N N N N 1 Tree Diagrams Method Y N N N N 1 Barrier Analysis Tool Y N N N N 1 Source: Reality Charting, Dean L. Gano Comparison of Root Causes Analysis Tools & Methods
- 24. RCA Apollo Method Example
- 25. Questions?