View
6
Download
0
Category
Preview:
Citation preview
Obtain real-world data, including: • Station inventory (configuration of
equipment, including any redundancy) • Equipment attributes (e.g. cost, reliability) • Support structure (e.g. shipping times)
The Challenge: The IMS sustainment mission is to ensure that IMS facilities continue providing data in accordance with availability, timeliness and other requirements as specified in the draft IMS Operational Manuals, at optimal cost.
[source: CTBT-PTS-INF.1163 Integrated Logistics Support For The International Monitoring System]
Minimum Requirements: ≥98% Data Availability at waveform stations (no more than 7 days down-time per year) ≥95% Data Availability at radionuclide stations (no more than 15 days down-time / 7 consecutive days per year)
[source: CTBT-WGB-TL-11,17/… Operational Manuals]
IMS Network: 170 Seismic Stations (50 Primary, 120 Auxiliary), 60 Infrasound Stations, 11 Hydroacoustic Stations, 80 Radionuclide Stations, 16 Radionuclide Laboratories, located in 89 states.
Run simulation in initial model
Example: Compare output results against measured Data Availability and other Key Performance Indicators.
Review MTBF (Mean Time Between Failure) values of modelled equipment,
and update values as necessary. Assumption: equipment failures follow Poisson distributions. MTBF values calculated with individuals confidence levels
Validated model can now be used to make decisions, such as: • Identify spare parts required and their
optimal location • Enable proactive decisions rather than
reactive interventions
Purpose of the displayed process Identify the optimal level of spare parts, and their location, required to sustain the IMS Network at the required level of Data Availability given a budgetary constraint. How ? Logistics Support Analysis provides the tools required to solve such problems. In this case the problem can be solved using a spare parts optimization model to simulate the IMS Network, using logistical data from the Database of the Technical Secretariat (DOTS) and from other sources to calculate the initial Operational Availability (Ao) of each modelled station in the Network. The spare parts required to meet the Ao equivalent to the desired target Availability can then be calculated, using an iterative marginal analysis process
Abstract The International Monitoring System (IMS) is to consist of 321 monitoring facilities, composed of four different technologies with a variety of designs and equipment types, deployed in a range of environments around the globe. Despite this, the network is expected to reach extremely high levels of data availability which could induce unbearable Logistics Support costs. The IMS is now already performing Logistics Support Analysis (LSA) enabling us to: identify the most efficient improvements to our Integrated Logistics Support (ILS) strategy, optimize sparing as early as the design phase of new stations or major upgrades, and explore alternative designs or maintenance policies. Initial results have already been obtained and have proven the benefit of such analysis. The results of such simulations will be instrumental in validating the Integrated Logistics Support (ILS) strategy, the Engineering Design and ultimately the overall Network effectiveness and capability.
DOTS Station Operator
Current Spares
Invest
ment
0.50
0.55
0.60
0.65
0.70
0.75
0.80
0.85
0.90
0.95
1.00
Ope
ratio
nal A
vaila
bilit
y
Cost
G Availability
G Peak Net Benefit
BRT Availability
BRT Peak Net Benefit
Current Spares Investment
G End Point
BRT End Point
G Net Benefit
BRT Net Benefit
20%
31%
22%
2%
0%
5%
10%
15%
20%
25%
30%
35%
Cos
t as %
of P
arts
Dep
loye
d
Support Costs
Spares Cost (BE)
Spares Cost (BRT)
Spares Cost (G)
Annual Demand
Logistics Support Analysis in practice: LSA is a great help to achieve high levels of operational
availability, reliability and/or safety. First introduced in the military aircraft industry, it is now used in the wider aerospace and military industries, and commercially where sustainability and reliability are paramount as
well as anywhere slight changes generate consequent financial impact.
• Optimized spares lists are compared to parts failures history
• Spares flagged as superfluous compared to failures history (or ‘lack of failure’)
Overstock
Optimal stock level
• Model stations Ao are compared to actual data availability figures
T3-P48
IMS Sustainment – Logistic Support Analysis (LSA) – How Theory Improves Reality N. Brely, JP. Gautier, D. Foster and D. Morozin Monitoring Facilities Support Section, International Monitoring System Division - Comprehensive Nuclear-Test-Ban Treaty Organization
Disclaimer: The views expressed on this poster are those of the authors and do not necessarily reflect the view of the CTBTO Preparatory Commission.
Logistic Support Analysis (LSA) provides a set of powerful tools to support the IMS sustainment mission as described within the ILS strategy. These tools are able to analyse large quantities of data and enable decision makers to turn historic data into predictive information, optimize the use of resources, and evaluate alternative scenarios. Such tools add to the intuitive understanding of the logistic support needs of the complex IMS network, which is not permitted to fail. Decisions made with this additional information are therefore stronger.
MFS has already: Identified the key drivers in terms of: Reliability of parts Cost of spare parts Cost of parts at recapitalization
Reviewed significant data elements of these key drivers, such as: MTBF Repair policy (location of repair, successful repair rate, cost impact of repair or consumption) Item cost
Provided results (using a reliable and repeatable method) to be used in financial planning, including: Annual budget for spare parts purchases Projected cost of station recapitalization
Produced an optimized list of spare parts for a set of IMS stations (and will soon be ready to extend the model to the whole IMS network) Identified an optimized relocation of existing spares, increasing theoretical Ao at no cost.
Next steps: • Roll out the optimization and redistribution of the sparing posture for the Australian subset of the IMS network, • Expand the scope of the model to encompass the full size of the network, • Analyse failures distributions (Poisson shape or not, constant across lives or not) and refine MTBFs, • Improve maturity of data in general, including support structures, shipping times, administrative lead times, • Benchmark data and results against industry best practices, • Explore “what if” scenarios…
Achievements & Way forward
Mathematical principle The calculation can take account of any existing investment in spare parts (‘brownfield’ analysis) or none (‘greenfield’ analysis), and uses an iterative marginal analysis process to identify the series of parts (and their location) which gives the optimal increase in Ao based on the cost, until all targets have been reached.
SnT 2013
Costs comparison: Installation Investment = 100% Spares Investment (G) = 22% Spares Investment (BRT) = 31% Initial Spare Investment (BE & BRT) = 20% Annual Demand = 1,5%
Validated model can also be used to investigate hypothetical situations, such as: • Sparing requirement of alternative
station design, • Impact of alternative support
structure, • “What if” scenarios • ….
Evolution of the Model
Outputs
Model size The results were generated with validated data from the Australian subset of the IMS Network. This subset was made up of: • 17 certified stations, • split within 3 Station Operators, • covering all 4 station technologies, • and a wide variety of locations (including remote island and
polar, as well as mainland locations) Amongst the 264 certified stations, this subset is 6.5% of the current network.
The wrong spare can create down-time !
Station Down Translated into CTBTO world Station Up
Key Terms • Operational Availability (Ao) is the probability that an item or system will operate
satisfactorily at a given point in time (here an average value across the whole network)
• Spares Investment is the initial investment required to meet the target Ao not including “Annual Demand” (spares renewal).
• Initial Spares Investment is the arbitrary initial investment provided in order to evaluate a configuration (actual or hypothetical)
• Net Benefit evaluates the decreasing return on investment as stock is purchased to meet the target Ao.
• Annual Demand is the consumption cost of maintaining this level of Ao (spares renewal)
DOTS DOTS
Calibrate model: • Validate simulation results • Review input data (e.g. MTBF, OSTs) • Validate new results (e.g. against KPIs) • Repeat as required…
Accuracy of the validated model is constantly measured and compared against actuals
Example: Use CoRe (Cost over Reliability) value of equipment to identify cost drivers and prioritize analysis:
𝐶𝐶𝐶𝐶 = 𝐶𝐶𝐶𝐶 × 𝑄𝑄𝑄𝑄𝐶𝑄𝐶𝑄
𝑀𝑀𝑀𝑀
4% 10%
1%
24%
38%
23%
MTBF review/by model
< 0.5
0.5 - 0.9
0.9 - 1.1
1.1 - 2
2 - 5
> 5
Prob
abili
ty
Running time until failure
Poisson Probability Distribution
Recommended