Control lists
Description
A control list includes the controls chosen by a user and can be applied to a dataset, a merged offer or an aggregated offer. The control list is used to detect errors left in the dataset.
View Control lists
You can consult the list of control sets created by your organisation in Workbench → Control lists.
For each set of controls, the following are displayed:
Name
Comments
If the control list is shared with the workgroup
If it is linked to a processing rule
The date of the last update
Creating a Control List
Add a control
Clicking on the "Add a control set" button takes you to the edit form for a new control list.
You need to assign a name to your control list. You can add a comment and decide whether or not this control is shared with the workgroup. If the control set is shared with the workgroup, other organisations that are members of your workgroup will be able to use this control list.
You must select a control to include and then click on "Add a new control".
You can add several controls to the same control list using the "Add a new control" button.
Depending on the control you choose, you will have to fill in the mandatory fields indicated by an "*".
You will always be able to add a description of the objective of the control, add a comment and choose the criticality of the control.
Add a context in the control list
You can also add a context to your control list. This means that the controls in the list are subject to one more condition.
Transport mode
You can create a control list with a "Transport mode" context.
Controls subject to this context will only run on models associated to the transport mode you have selected. Models with other transport modes will be ignored.
Once the context has been created, you add the controls in the same way as shown above. These controls will be subject to the defined context.
Next operating period
You can create a set of controls with a "Next operating period" context.
With this context, controls will only run on models operating on the selected days.
For example, you can check the presence of an attribute on lines operating on the next 30 days. Since the context is "operating on the next 30 days", lines not operating on the next 30 days will not be checked for the presence of an attribute.
Types of control
Presence of an attribute
With the "Presence of an attribute" control, the user can check that a specific attribute has been filled in on the selected model.
The user can check all the non-mandatory attributes of the following targets:
lines
stop areas
vehicle journeys
journey pattern
company
For example, the user can check the presence of the "Associated web page" attribute in the line target model:
Presence of a code
With the "Presence of a code" control, the user can check that a specific code has been entered in the selected model.
The user can select any code space defined in the Workgroup. (see Code Spaces )
The models targeted are :
lines
stop areas
vehicle journeys
For example, the user can check whether the "ticketing" code is present in the target vehicle journey model:
Presence of a custom field
If one of your repositories contains defined custom fields, you can check for their presence using the "Presence of a custom field" check.
The target models for your control are as follows:
lines
stop areas
company
journey pattern
vehicle journey
Code format
The user can check that a code has been entered in a specific format. This control can be performed on the following models:
Stop areas
Lines
Vehicle Journeys
Models with no code are ignored. Only models with the selected code are checked.
The expected format of the code must be specified using a regular expression.
Below are some typical examples of regular expressions:
begins with a word: ^word
ends with a word: word$
contains a word: (word1|word2|word3)
Format of attribute
The user can control whether an attribute in a model is filled in with a specific format. This control can be run on the following models:
Stop areas
Lines
Vehicle pattern
Vehicle journey
Company
You must select a target attribute, which will be subject to the control, present in the chosen model.
The expected format of the attribute must be specified using a regular expression.
starts with: ^word
ends by : word$
contains word : (word1|word2|word3)
Service count trend
The user can control that each line has a similar number of services each day to the previous and following weeks. The number of daily services is compared with the number of services on the same day of the week (Monday with previous and following Mondays).
The user can define :
the number N of weeks preceding the day analysed
the number M of weeks following the day analysed
the maximum difference P in the number of services (as a percentage)
the level of criticality raised by the check (error or warning).
When the control is run, an average is calculated including the same day of the week over N weeks before and M weeks after. If the number of services deviates from the average by more than P%, a warning or error will be raised by the control.
Speed into Journey Patterns
Users can control speeds within Journey patterns. A minimum and maximum speed and a minimum distance are to be defined in the control.
An error or warning will be displayed if the control detects a value in excess of those defined.
Presence of associated models
The user can control the presence of models associated with a target model. A minimum or maximum number can be specified for each model.
An error or warning will be raised if a value in excess of those defined is detected by the check.
Model status
The user can control the state of a model. A warning or error will be displayed if the state (enabled/disabled) of the target model does not conform to the state expected by the control.
Inclusion in a geographical area
The user can control whether a model is included in a given geographical area. In this way, a warning or error will be displayed if the model checked is not present in the defined zone.
For each model, you need to enter the coordinates of the top left-hand corner and the bottom right-hand corner. These two points are used to define a rectangle forming the perimeter of the geographical area to be checked.
Expected provider
Users can control if the provider of a model belongs to their Workbench.
If the supplier does not belong to the user's Workspace, an error or warning will be returned on the report.