MTBF-MTTR in process flow

[yoni_a]

[ FlexSim 18.1.1 ]

How can I incorporate the logic of uptime and downtime in a process flow, similar to the logic that is added to the processor in the 3D view. and example would help

[emily_hardySTLSK · Accepted solution]

This depends on whether or not you want the breakdown to apply to 3D models or to a token.

If you are using a 3D model and want to use process flow to create a breakdown, create an interarrival source with the breakdown data, and then create a Custom Code activity that stops the object. Then you can add a delay, or add a sequence of activities that the operator will follow. Then, create a second Custom Code to resume the objects.

If you want the breakdown to apply to the tokens, wrap your process flow in a zone. Then create a block like this to the side of your process flow:

This will save the tokens position, release the token, and then restore it to its last location when the downtime is over.

If you let me know which of these options you were looking for, I can create a sample model for you. Let me know if this helps!

[yoni_a]

thanks @Emily Hardy

[yoni_a]

I would like to apply it to a token @Emily Hardy, would you use the same block for breaks and lunches downtime will that be a separate block?

[emily_hardySTLSK]

For breaks and lunches I would create a separate block as a scheduled source, and create a table of the scheduled breaks and lunches. The one I used above will be created at a random time.

[yoni_a]

Hi @Emily Hardy, can you send me the sample model for both the MTTBF and the break and lunches

thanks

[emily_hardySTLSK]

Here's a model showing these two interruptions.

yoni-support-model.fsm

[yoni_a]

hi @Emily Hardy, thanks for your model,

I am trying to understand your model; your model you have a resource (QTY 1); with a cycle time of duniform(1, 10), * why do you put it inside a Zone?

Lunch and Breaks:

You have a schedule source, I assume that between 12 and 13 you have the first break and from 16:00 to 17:00 the second, but you see the first token generated at 12:35 and the resource (delay-token) is filling up which means it is not on break, the second token for the break is generated at 16:23, and you have the same condition. a token is filled within the zone, when if it is on break it should stop

Breakdowns

you have a source with exponential (0,200) I assume this is the distribution for your downtime, then you have a delay exponential (0,10), what does the delays do? again I assume each token generated by the breakdown is a random downtime, but when it is created the token in the resource-delay block keeps filling up, which I assume it did not go in a breakdown mode

waiting area.

what is the logic or reason for the waiting area?

how do you connect the breakdown and lunch blocks with the resource activity, I would assume if you generate a token in the breakdown and breaks , the token in the zone-resource-delay will have to stop as it occurs in the 3D model.

How do you simulate the uptime? the box only shows downtime?

I am having a hard time understanding how it works?, how do you reference several breakdowns for individual machines if you have several?, I don't see how it references a specific resource.

thanks

mtbf-mttr3.png

[emily_hardySTLSK]

The zone is the area where you will have the breakdown (think of it like a processor).

The delay in the breakdown represents the time it takes to repair the machine. You can replace this delay with something else if you prefer, like a repair sequence or something. In the lunch time, the delay represents the time of the break. That is why the tokens fill up.

The waiting area is just where the tokens are released to upon breakdown. There is no code associated with it, it's just a place for them to wait and then they will be taken out when the breakdown is over.

Uptime is any time that is not downtime; you don't need a separate activity for that.

Does this make more sense?

[yoni_a]

thanks @Emily Hardy,

1-I understand the delay part on the breakdown and breaks, how it is saved and later release to the waiting area and then restored. The part that I don't get yet, is that you use three separate sources, and I see that the token fill up in the breakdown box and the breaks, but if the delay in both represent a downtime, why does the processor keeps filling up as if it is processing

2-lets say I add 3 more processors how does the breakdown and box lunch block, knows to which processor is belongs to? how can I get statistics of the total time of downtime due to break and lunch?

3-I assume the source at the beginning of breakdown block is the random variable, but in my case I don't have the timing or frequency of the downtime, I only have the day it went down and for how long it went down

[yoni_a]

Hi, @Emily Hardy, in your lunch your schedule source have a 1 from12 to 13 and from 16 to 17, but the token is generated at 12:25 and 16:23 why? shouldn't it be generated at 12 and 16?

[emily_hardySTLSK]

Sorry- I meant to change one of the parameters. Currently FlexSim will create a token randomly between 12 and 13. Change the "Arrival Spacing" to "At scheduled time" as shown in the screenshot.

[yoni_a]

@Emily Hardy , see previous reply for this