Unlocking Advanced RAM Modelling with Availability Workbench

Recently we hosted an exclusive webinar that explored advanced techniques for Reliability, Availability, and Maintainability (RAM) modeling. Attendees had the chance to go beyond the basics of RBD logic, Mean Time to Failure (MTTF), and Mean Time to Repair (MTTR) and dive into the powerful capabilities of simulation techniques using Availability Workbench.

What We Covered:

• Beyond the Basics: We discussed the limitations of simple metrics like MTTF and MTTR and how to tackle real-world challenges effectively.
• Monte Carlo Simulation: Attendees learned how this technique can predict system behavior under uncertainty and model complex scenarios with confidence.
• Real-World Applications: We demonstrated how to incorporate wear-out failures, maintenance schedules, and rotating equipment into RAM models.
• Optimization Strategies: The session showcased how advanced modeling can enhance both reliability and maintenance strategies.

Key Takeaways:

The webinar provided practical insights for engineers, analysts, and managers who:

• Want to build more sophisticated and accurate RAM models.
• Seek to optimize reliability and maintenance planning with data-driven approaches.
• Are new to Monte Carlo Simulation and curious about its applications in RAM modeling.

Looking Ahead:

If you missed the session, don’t worry! We’ll be sharing more opportunities to learn about advanced RAM modeling techniques in the near future. Stay tuned for updates and announcements.

Stay Connected:

To ensure you don’t miss future events, subscribe to our newsletter: Home Page - Isograph or follow us on: (30) PeakAvenue GmbH: Posts | LinkedIn .

The event was a success, and we’re excited to continue helping professionals unlock the full potential of RAM modeling with Availability Workbench.

Included is a list of questions and answers from this webinar:

what are the differences between MTTF and MTBF? i usually input MTBF for failure data.	MTTF (mean time to failure) is the average time between start up (or installation) of a new asset and the point of first failure. MTBF (mean time between failure) is the average time from one failure to the next. Strictly speaking, MTBF = MTTF + MTTR. MTTF and MTBF are often very similar as MTTF >> MTTR. This may be why the terms are sometimes used interchangeably.
how does the number of intervals work and what's its implication on the output result	The number of intervals has two functions. 1) It is the number of time increments in the profile plots. For example, if you set a lifetime of 10 years and 10 intervals, the program will plot system parameters like cost, downtime, etc, on a plot with 10, 1-year time intervals on the x axis. 2) It is the interval used when calculating net present value (NPV) and cost escalation. In the settings, you can specify a NPV yield % to be discounted from the system cost and an escalation % to be added to various other costs (e.g. labor, spares, etc). These % discounts/escalations will be applied per interval.
i was thinking, sometimes supplier only provided us lifetime for a product, can i treat it as MTTF and input to the RBD? or as what Jonathan mentioned, it's the same as MTBF in Isograph program?	It may be best to consult with the supplier about their exact meaning. Is it the recommended time to replace the component, or is it the average expected time of failure?
There was a possibility to "seed" the random number generator	You may set a random number seed. This acts as a trigger for the random number generator. It is an integer. The same sequence of "random" numbers will always be returned as long as the seed is not changed. This is important as it means that any change you observe in your results will be due to a change in your model, and not because of a different sequence of random numbers.
Isn't the tank just an additional cold standby redundancy with a very short life time?	Temporary backup may be modeled using a Buffer distribution. When defining a failure model, you may select Buffer as the failure distribution (same place where I selected Bi-Weibull). Now you will be able to set Time to Fill and Time to Empty. Placing the block in parallel with a pump block for example would provide back up for the TTE specified. There are also different regimes for how the buffer is refilled, e.g. only refill after pump is repaired, change refill rate based on capacity available, etc.
Would you say that AvSim is a combination of the module RCMCost and RBD?	In a sense, yes. The failure models are very similar to causes in RCMCost, and are simulated in the same way. Combining ideas like RBDs and rules allows us to model the dependency between failures in a way that is not possible in RCMCost.
1) How AVsim handles multiple products? (where there is equipment that handles fluid A or B in the same RBD).	In the Project Options there is a Products tab. Here you may define up to 4 different product streams. The program can then perform capacity analysis for all the products defined.
(2) Does the process relationship of fluid A with respect to fluid B have to be included, or does everything refer to 100% of each fluid separately?	Each fluid would be treated independently. For example, if a block can carry 50% or A but only 20% of B, the loss of that block (assuming no redundancy) would result in the loss of the specified amount of each product.
6 months maintenance then good as new, with a 6 months wear out, cancels out no?	It does! I chose a simple example to illustrate clearly the relationship between maintenance and wear out. Naturally it may not be so simple in the real world, especially when balancing costs and losses due to maintenance against those due to failure.
Does the equipment rotation rule take into account planned maintenance as well as component failures?	Yes, a PM would trigger rotation, but only if the PM causes the block to become non-operational. This is a setting in the properties of the task where you can tell the program if the asset needs to be shut down for maintenance or if it can remain operational during the task.
Have you also done the analysis with as good as old?	Yes. This results in a substantial increase in downtime. In fact the result is similar to the model without PM since the PM now has no impact on the condition of the system. (there is a slight difference owing to a change in sequence of events)
Is there a way to find the output results for a specific Phase only, and disregarding the downtime during the other Phases?	Phase results are limited to costs only. However, This could be achieved using the profile intervals discussed in an earlier answer. The program will store downtime for each interval, allowing you to output a custom report that omits the downtime from one or more intervals if you wish.
Could you explain a case for Availability vs. Production Capacity using AVsim? How can I activate it in my AVsim?	It would depend on what is required from the model. A straightforward system availability analysis would look only at downtime, i.e. only instances of a complete outage of the system would be considered. A capacity analysis would be useful where it is necessary to consider degradation of a system, i.e. failures that do not necessarily bring down the whole system, but impact its ability to operate at maximum production. To activate a capacity analysis, create a Consequence (Add >> Resource >> Consequence) and select the Loss of Capacity option in the properties dialog. Next, double click the system-level block in the project tree and select the new consequence from the Consequence drop-down box. Now, when you run the analysis, the results will include the average lifetime capacity. It will also be possible to produce a capacity profile plot report. Note that you can adjust the max capacity of each block by double clicking and selecting the Rules tab.
Is the integration with EAM tools already in place or are you planning to do that?	Integration is available for SAP and Maximo. These portals primarily work with the RCMCost module, as it can accept asset hierarchies of the kind stored in such tools. However, AvSim can still make use of other data downloaded from an EAM, such as spares, labor, work history and maintenance strategies. If you are a user of RCMCost as well, the asset hierarchy download from the EAM system could be converted to a simple RBD using an option in the Tools menu.
How can you control when a repair task is done after an inspection finds a failure? I have seen it be repaired based on when it would have failed but in the real world you would like to control when it is repaired.	At present there is no means to specify exactly when the preventative task would happen. The reason the secondary task is deferred until the point when the failure would have happened is that this allows the program to determine if there is enough time available to requisition the resources needed for the preventative task. Say, for example, the PM requires a spare, but it would take a week to get the spare and the failure is set to happen after 5 days, the program would advance the clock to the point of failure, find that the resources have not arrived and trigger a failure event instead.
How can we add the backup time of a tank by available liquid level (example 4 hours) in the event of a pump failure?	Please see previous answer describing Buffer models.
(a) ¿Cómo AVsim maneja varios productos? (ejemplo algunos equipos con fluido A y otros equipos con fluido B en el mismo RBD). / (b) ¿o es necesario adicionar la relación entre la capacidad máxima de los fluidos?	Please see previous answer regarding different product streams.
Please comment a case about Availability Vs. Production Availability using AVsim? Could you clarify if it is activated at Level 1 (more general) by adding a particular block or if a selection is activated at the Project Level of the options menu?	Please see previous answer regarding activating capacity analysis in AvSim.
How do we incorporate the Ramp Time into our model?	Ramp time may be added in the properties or each corrective and planned maintenance task. It is immediately below the Task Duration field in the Task Properties dialog.

If you have any questions, please feel free to contact me.

Best Regards, Jeremy