Resource Scheduler - Hour-level timescale

Related products: Software Factory Universal GUI

We use the Resource Scheduler in the Universal GUI to plan our Vessel fleet. For Operational purposes we need to be able to plan and display multiple Activities within a Day. Therefore we need an Hour-level timescale and the Resource Scheduler to place Activities after each other when Activity 1 datetime end is before Activity 2 datetime start. Currently they are stacked when the Day matches, even if the Time is not overlapping.

Can this functionality be added to the Resource Scheduler please?

This doesn't work for us

 

Great idea. Would be very handy indeed. 


Updated idea statusNewOpen

Good to see Arie’s suggestion being turned in an "open” idea 😀 Instead of an hour based scale, we would like to take it further.  Can it be a variable scale? We could use a scale of 15-minute time blocks for a client.


The current scheduler in Universal works heuristic - you cannot cannot yet configure the scheduler in the model. 

Universal attempts to set up the scheduler as well as possible using the information present in the model. The from- and to- date are automatically picked based on column order. The display value of the row is used for the row, conditional layouts on this display value are applied and the grouping on the left is based on the configured grid grouping.

However, there are no settings available to derive the desired timescales from. We may be able to double-down on the heuristic approach by deriving the available timescales from the precision of the dates - when it is a DATE it allows year/month/week and year/week/day and when it is a DATETIME it allows for week/day/hour timescale as well.

To eventually replace the heuristic approach, we plan on formalizing the scheduler settings in the model. This is currently still in the design phase, but the idea is to allow the developer to configure the available sets of timescales, enabling them to have certain intervals (year/quarter/month/week/day/hour/quarter-hour/minute).

The following things are being considered:

The most coarse timescale will likely always determine the ‘page’ size of the scheduler. For instance, when the top timescale is year, it will allow you to page year-by-year. Since this may cause some difficulties at the boundaries (around new years) we might opt for navigating half the top level timescale on page navigation.

The most detailed enabled timescale will determine the cell interval. This may be quarter-hours, hours, days or even weeks and will determine the amount of time that is modified when drag-drop is performed.

Any enabled timescale in between will allow for extra header labels.

If possible, we’ll allow the option to omit headers for the lowest timescale. This will allow you to set up a week/day/hour/quarter-hour timescale, using the week as pages, day and hour as header labels but having the cells on quarter-hour level so you can more precisely drag-drop without cluttering the header labels with quarter-hours. This will also prevent the overlapping mentioned in the first post of this idea.

Also, if possible we’ll allow the developer to set an interval for each timescale, allowing you to use 2-year pages or configure a minute-based scale as 10-min, 15-min, 30-min. We are still in the process of working out whether or not it should always fit the preceding timescale. Having a 26-min timescale under an hour-timescale would be a bit odd.

 

We would love to hear your input on this, both the formalization in the model as well as the heuristic approach for allowing additional timescales until the formalization.


Hi Anne,

I only just now saw your reply to this topic from a month ago. As I understand it, you guys will be working on this in the near future. I'm not sure who's input you ask for, but I can give mine. Reading your post, I see that you have two opions that you consider implementing. The first being the existing heuristic one, with an extra level of depth to hour, against the current deepest level of day. The other option is more complex, but offers deeper levels, up to minutes even.

That second option is approaching what we need. We need a planner that supports the following functionalities:

  • The planner can schedule units with a minimal duration of 5 minutes. Maybe it would even be necessary to be able to schedule per individual minute. Preferably, the scale will be configurable, so that we can go as low as 1 minute of time, and scaling up if necessary. We need varying units of time.
  • Units of time can visually be dragged across the planner to reschedule.
  • Units of time can visually be resized to lengthen or shorten the time of the unit.
  • To the left of the visual planner, we need two columns. The first column (left) contains a main entity, the second column (right) contains sub entities that are part of the main entity, or that relate to the main entity. Both columns can be sorted ascending or descending by clicking their headers.
  • The current scheduler can switch between days and weeks. Our scheduler would need a switch for 5 minutes and a multiplication of that amount.
  • The scheduler needs to support colors for the visual scheduled units of time.
  • As we need small units of time, the schedulers probably will take up more horizontal space than the available screen estate offers. Therefor, the scheduler will have a scrollable visual section, just as it has now. Of course, this works for vertical visualization as well.
  • Currently, the visualization uses squares as a visible unit of time, having a value of one (1) day. We need smaller units of time, so the square will need to be narrowed down to a rectangle in order to keep the planner from growing sideways too much. However, the rectangle cannot be too narrow, because users will need to be able to drag and resize the rectangle if they need to.
  • A clear indication of the current moment in the scale at the top of the scheduler.
  • The scheduler can be localized/translated.
  • Open the planner at a configured date and time, not at the current date and time. Only if no date and time to open the planner at is configured, the planner can open at the current date and time.

Against the list above, the second option you elaborate on would be preferable to us.

Best regards,
- Alex


Hi Anne,

I only just now saw your reply to this topic from a month ago. As I understand it, you guys will be working on this in the near future. I'm not sure who's input you ask for, but I can give mine. Reading your post, I see that you have two opions that you consider implementing. The first being the existing heuristic one, with an extra level of depth to hour, against the current deepest level of day. The other option is more complex, but offers deeper levels, up to minutes even.

That second option is approaching what we need. We need a planner that supports the following functionalities:

  • The planner can schedule units with a minimal duration of 5 minutes. Maybe it would even be necessary to be able to schedule per individual minute. Preferably, the scale will be configurable, so that we can go as low as 1 minute of time, and scaling up if necessary. We need varying units of time.
  • Units of time can visually be dragged across the planner to reschedule.
  • Units of time can visually be resized to lengthen or shorten the time of the unit.
  • To the left of the visual planner, we need two columns. The first column (left) contains a main entity, the second column (right) contains sub entities that are part of the main entity, or that relate to the main entity. Both columns can be sorted ascending or descending by clicking their headers.
  • The current scheduler can switch between days and weeks. Our scheduler would need a switch for 5 minutes and a multiplication of that amount.
  • The scheduler needs to support colors for the visual scheduled units of time.
  • As we need small units of time, the schedulers probably will take up more horizontal space than the available screen estate offers. Therefor, the scheduler will have a scrollable visual section, just as it has now. Of course, this works for vertical visualization as well.
  • Currently, the visualization uses squares as a visible unit of time, having a value of one (1) day. We need smaller units of time, so the square will need to be narrowed down to a rectangle in order to keep the planner from growing sideways too much. However, the rectangle cannot be too narrow, because users will need to be able to drag and resize the rectangle if they need to.
  • A clear indication of the current moment in the scale at the top of the scheduler.
  • The scheduler can be localized/translated.
  • Open the planner at a configured date and time, not at the current date and time. Only if no date and time to open the planner at is configured, the planner can open at the current date and time.

Against the list above, the second option you elaborate on would be preferable to us.

Best regards,
- Alex

This list is already pretty extensive and complete. I would like to see most of it as well. Adding some more UI elements:

  • conditional formatting on day characteristics.. such as weekends / holidays
  • dependency lines, to visualize dependencies between planned entities. 
  • tooltips (hover) that allow for more information to popup, which preferably can contain a html field where deep links to the application can be shown. 

 


Good additions, Freddy! 👍


OpenPlanned

Hi all,

It seems that this Idea gets overloaded with a lot of (great) ideas. To keep the focus on this Idea, I recommend you all create separate ideas for them. That way we can see the different interest in them.

This idea "Hour-level timescale” is planned for 2023.1. We intend to implement not only Hour timescales but the following: Year / Quarter / Month / Week / Day / Hour / Minute

 

 


PlannedWorking on it!

In combination with platform version 2023.1, custom timescale support is now supported in the latest releases Universal GUI (2023.1.12). This version already available as a beta release now and will be officially released next Friday. 

Note: there are a lot of combinations of configurations possible. We've tested of lot of them and notice not all are shown properly. So, choose your combination of timescales wisely.

The obvious ones will work, and we focused specially ones the ones requested here.

We will close this idea now and feel free to let us know your findings.


Working on it!Completed