I am modeling a robot that picks up and delivers items in a warehouse(on the moon). One of the entities is the box the item is stored in, this entinty has an activity “store and protect item”. This activity starts before the scenario and continues even after certain interactions ( the box is picked and moved with the item still in it).
I understand that the vertical yellow bar shows the time of the activity, although I cannot drag it above it’s first interaction. I am not sure how to show that this activity was already ongoing before the scenario ad continues after interactions.
Just a thought, but have you considered using modes and states to model this instead? I.e. the box has the state “full” and “empty” that it alternates between?
Hi
From what you say I understand the box is exchanged from robot to warehouse: would it be an exchanged item instead?
Then the robot gets items from context, and later delivers filled box to warehouse. Then the found item refers to a class describing what is the item, the box refers to another class which owns the one of the item with a cardinal 0 to some. What is missing is what happens to empty boxes: how the robot get some to continue the collection.
To me the story of the box can generate another model, pending if you need separated system design or if direct product design can do the performance dimensioning…
Hope this helps
Thierry
Edit1 : The different EIs there are different states of same objects : item alone, empty box, filled box. For the box one of the class property has to be contained item count on which you can base a state. Classes may have state machines even.
I did not think of this, thank you!
That’s very helpful, thank you. I am stuck in the system analysis and was not sure if I should be representing these concepts at this level
I have a similar issue where I’m trying to represent the environment presenting weather and obstacles to an object detection system. The environment is constantly providing the weather conditions under which the obstacle detection system must recognise (to report on its performance, e.g., visibility range) and potential objects that could be the cause of collisions (that the object detection system must recognise). At the moment I’m using PAR with “provide object kinematics” and provide “weather conditions” as activities one segment, and the other object detection activities in the other segment. Is this an appropriate way or is there something better? Attached is an earlier iteration of the scenario:
