How to find the idle time of the job/product?

Hello

In my model, i have specified the arrival schedule of the jobs and the processing times(load+process+unload time).. At the end of simulation i'm able to get the idle time and the processing time for machine/processor...
But if we take the case of a single job, Is there any way to find out the idle time/waiting time of a job in queue?


Also, we have collected the waiting time of 30 jobs from the real system (There are about 26 machines and we have followed every job/product on these machines. Waiting time of each job i.e., the time a job required to move on to next machine, has been collected ).
Now, Can this data(waitin time) be used in the model? If yes , how and where to enter this data in the model?

Yes, in "Flexsim Charts" )the post run reporting tool in Flexsim) you can obtain flowitem reports that include a gantt chart of the states that a particular flowitem transitioned through in the model. Go to Statistics>Reports and Statistics menu. You will need to have history turned on before running your model. If you're new to Flexsim Charts, you might want to read up on it in the help first.

Thanx Cliff for ur help...

Few more problems in my model...

Every time i run the model for a specific amount of time, it gives constant result... like 1st day- 166 jobs/products; 2nd day-346products, third day-526 and so on....
This is true for any given time i.e,. if i stop the simulation and rerun it for any no. of days then also the result is same(it is adding 180 parts/products per day)...

Regarding the inputs...


There are 26 machines.. at the beginning of day1 200 jobs are brought in and as sooon as the stock is over, another 200 will come into picture...
Iam using repeat arrival schedule for this step...
The processing time(setup, processing and unload time) of all the machines for each job/product is constant like 1st machine 321sec, 2nd machine 317sec and so on........
What could be the reason for constant output? Is it the constant source flow and processing times? how to obtain variation in the output?

Next we have collected the waiting time of each job i.e., the transition time for moving the job from 1st to 2nd machine and so on... this time is quite varying where as in my model i have made used of conveyors for the movement of jobs which take constant time...All the conveyors are taking a constat time of 2seconds to move the part from one m.c to another whereas the real data shows it varies between 2 sec to 15 seconds... How to make these changes?
Is it possible to incorporate this data(waiting time of job) in the model and run it?


In real situation about 190 jpbs are produced whereas in my simulation, it is constant 166jos on 1st day and thereafter 180 jobs are added each day...
I am attaching the screenshots of my model

Shivrash,
If you are using constant times throughout a model then you will always end up with the same result each time you run the model. Adding variable times using distributions instead of constant times would make the output of the model vary from run to run and over time.

If you want to add a variable travel time between locations in you model then the best way for you to do that at this time would be to use a processor to simulate the travel time and set the process time to use a distribution that matches the data that you have collected from the real world.

If you are producing less parts per day in the model at this point in time there are two possible reasons that I can think of that would cause this. First, your constant travel times may be longer than the average variable time in the real world. Second, your process times in the simulation may be longer than they are in real life.

Not having seen your model or the real process I can only guess but it sounds like your simulation may not be set up right. There is a correlation between the level of detail that you put into a model and the level of detail that you get out in the data. It sounds like you are simulating the movement of the parts differently than in real life. A more detailed description of the process might enable us to help you more.

Brandon

Thanx Brendon for your valuable suggestions...

I would like to explain the situation in detail...

It is a continuous production process. There are 26 machines in a Z type of layout....They bring in 200 products in the stock for machining..Before this stock of 200 is over, another stock of 200 is brought in so that the production does not stop... Then the time for machining has been collected from the real system which is almost constant.. If we take the case of Ist machine, it takes about the same time for machining 2nd, 3rd & 4th job as it takes for the first job....
I Job II Job III Job IV Job & so on
I m/c 310 311 310 310
II m/c 85 85 85 85
IIIm/c 180 181 182 180

Above mentioned machining(load+processing+unload) times are in seconds... This data is almost constant, hence I have used a constant process times for each individual m/c...
Is it alright or Is there any way to give this data in form of distribution?
Next coming to the waiting time of job, there are push conveyors between each machine wherein the operator places the job on the conveyor and pushes it... this travel time is almost constant varying between 3 & 4 seconds... The important point here is , if we take the case of first few jobs, there may not be any waiting time for them, it is simply the travel time but the later jobs may have to wait in line for their turn to come on the machine(for these products , waiting time = travel time+waiting time) these waiting times have been collected from the system...
this is the only data which varies to a larger extent..
For ex: if we take the case of some Nth job then it takes the following waiting times..

Ist to II m/c 11sec
II to III m/c 10sec
III to IV sec 8sec
IV to V m/c 15 sec
V to Vi m/c 18 sec
and so on

In the simulation model, i have made use of conveyors which takes up a constant time of 2 sec for travel...


I am not able to understand how to make use of the above data or how to incorporate it in the simulation?

I am attaching the copy of my model..

Please help

Shivrash,

I looked at your model and have a number of comments on it.

• It looks like your missing a conveyor before the Bearing Cap Fixing Processor.
• All of the parts go through all of the machines in the model and the machines all have different process times. Ultimately your model will back up prior to the machine with the longest process time until it backs up to the source. Also, you will never have any waiting time after that machine because parts come out of the machine spaced too far apart for that to happen. If you watch your model run this happens with the Bearing Cap Fixing processor.
• If the process times for the machines were allowed to be variable then the waiting times between machines would be variable. NOTE: This would only happen in the absence of the above problem
• If it takes 3 to 4 seconds for a part to travel a conveyor to the next machine then your conveyors are either not long enough or their speed needs to be reduced.

It seems to me that the actual plant must run differently than the model is running. I'm just guessing, but if the times in the model are correct then I would bet that either some of the parts are skipping some of the machines or there is some kind of man power restraint between machines in the real plant. Ideally a process like this would run the smoothest if all of the processing times were almost equal, yet your times in the model range from as low as 60 to as high as 440 seconds.

At this point I would double check the actual process and make sure that I had accounted for all of the different factors that come into play with the processing of the parts. You have suggested that the model is not duplicating the real results and from this it seems as if there are some aspects of the process missing from the model.

Brandon