Hi.
I am new at modelling using Arcadia and I am wondering if it is a shift from the methodology to explicitly represent the system under study directly at operational level.
Indeed, I am representing a human remote-controlled system. The human operator thus can send commands which will be relayed by the remote system to the controlled system (both systems being part of the system under design).
I am asking this because by representing the remote-controlled system as an operational entity does not fit with the definition of an operational entity and it seems not to fit with the information I have red from there to there. In our context, it is not impossible to have such a “remote” concept directly expressed in the requirements.
What is confusing me is that it is possible to make the system (at system analysis) realize an operational entity.
Thanks for your time and answers.
Arnaud
Hi Arnaud,
it is important to understand what operational analysis (OA) is and when it is involved in engineering. OA is particularly useful when trying to best satisfy a customer need, without having an imposed system perimeter, or when seeking innovative ways to satisfy this customer need.
The OA should not mention the system because it aims to understand the customer (and other stakeholders) need, without any a priori on how the system will contribute to it; this in order not to restrict too quickly the field of possibilities.
If you don’t mention the system in OA, then you have the ability to later define several alternatives of how the system might contribute to stakeholders need, in several System Need Analyses (SA), each of them defining system functions and actors (incl. users) functions contributing to each operational activity, and to compare these alternatives. As soon as you put the system on the scene, then this game is over.
As an example, let us consider the operational need to hang a painting on a wall. If you already consider the system (let’s say a drill, crews and dowels) in OA, you will probably describe the activities to drill a hole and so, but you will miss possible alternatives such as using a hammer and a nail (which is easier and might be more appropriate depending on the wall material), or using a self adhesive hook (which is faster), or sticking the painting on the wall, which is also faster, more robust than the hook, better to avoid painting rotation… but not allowing removing it from the wall.
In this case, it is useful to allow for several alternatives to be considered; but even more important is to capture constraints that would not appear when system-focused: should it be possible to remove the painting from the wall in the future? Should its orientation be corrected so as to ensure strict horizontal positioning? How is wall material important? These are some major questions that the “systemless” OA should raise, and that you would probably miss if system-focused.
In your case, you might have an operational entity which would be the user, and describe his/her own activities, no matter he/she does it by his/her own, or thanks to system capabilities. Then in SA, you decide which part of these activities the system will contribute to, and how, thanks to defining system-allocated functions (and user-allocated ones); then you link each system function to each operational activity to which it will contribute. So yes, the system can (partially here) realize an operational entity.
Jean-Luc
Hi Jean-Luc
Thanks for this clear answer.
Arnaud