#11 How to model modes and their parameters

[Martin]

Let us assume a device having three different modes that can (and most likely will) be selected during the surgery:
a) normal working mode
b) working mode for special situation 1
c) working mode for special situation 2

Only one mode a, b, and c can be selected at any point of time.
The mode may be manufacturer specific.

To make things easy, let us assume that there are two parameters influences by these three modes:

• parameter x
• parameter y

What I like to model is which different parameters values are active and which parameter values will be active if another mode is selected. The parameter values shall be configurable no matter whether the mode is selected or not.

For example:
a) normal mode: pressure = 15mmHg; flow = 10l/min
b) high flow mode: pressure = 16mmHg; flow = 20 l/min
c) high pressure mode: pressure = 25mmHg; flow = 8l/min

In my opinion, manufacturer-specific, proprietary codes shall be avoided as much as possible.

Consequently, I see the following design questions:

• how to model the operating modes
• how to model the parameters
• how to model the link between operating modes and parameters

[Martin]

My idea

• modelling the operating modes as three different setting metrics with corresponding codes and corresponding (standardized) value set {ON, OFF/STAND-BY}. [right column in the figure]
• modelling one setting metric for each parameter x and y with codes describing the parameters (target flow and target pressure) [green boxes in the figure]
• modelling three pre-setting metrics for each parameter x and y for the three modes with the same codes as the setting metrics (target flow and target pressure) [dashed boxes in white, grey and dark grey]
• modelling two relations from each operating mode setting metric to the corresponding presetting metrics describing the connection "belonging together" (shall be standardized) [blue arrows]
• modelling a relation from each operating mode setting metric to the setting metric describing that if the mode is selected, these two metrics may be affected [black arrows]

Advantages:

• we can clearly identify which parameters belong to which mode
• the setting metrics to not need any proprietary codes, even of there are modes being proprietary

Disadvantages:

• having on/off-metrics
• lots of metrics
• having different setting metrics for three modes where only one mode can be selected. However, if we take one metric with value set {normal mode; working mode for special situation 1; working mode for special situation 2} I do not see how we can model the link to the parameters of each mode.

[Martin]

Idea to cope with disadvantage "having on/off-metrics"

Group the pre-settings into channels and use the activation state of the channel to model the mode on/off

** disadvantages **

• lots of channels
• "problem" of proprietary codes is shifted from metric to channel layer

[Björn Andersen]

I read the whole ticket again and I am still 100% convinced that this is exactly what we intended to solve with the "program" pattern.

@martinuro: Please elaborate on the difference (or we can schedule a call).