Enhancement Requests for Analytica


Key for Level of Importance

  • Would be nice *
  • Valuable **
  • Critical ***
  • I can't use Analytica without it ****
  • I want it so much, I'd pay Lumina to develop it! ****$

Requests

  • PageUp/PageDown shortcut keys to scroll along the X-axis of charts while maintaining the current axis scale. ** OR a full fledged scroll bar. ***

Bob Wimbrow, bwimbrow@sunedison.com

  • I've been asking for some time to get improved printing functions. In order to make the models peer-reviewable, there needs to be a simple way to print the entire model in a compact form with a list of each node, its formula, the nodes on which it depends and supports, its indices. Right now the options are really limited and when I print to a PDF it tries to create a new file for every node. [Evan Reis, reis@certuscorp.net]
  • An Undo History function, or at least the possibility of undoing more than just the most recent edit (as it seems to be at present). **** Jgrellier 01:43, 26 November 2007 (PST)
  • When viewing a graph and altering its Uncertainty Options it would be convenient if there was a Preview button, which would allow the user to see the result of changing these parameters. * Jgrellier 09:59, 26 November 2007 (PST) In fact, it would be nice if the Uncertainty options were integrated as one or more extra tabs in the Graph setup options, so you can see the results with Preview. [MH]
  • Ability to add Data Point Labels within a chart. This is specifically desired for creating tornado charts where each low/high point is labeled with the input value, but is also useful for many other types of plots.
    • If you click on any bar or line in a graph, it shows a popup window with its value (and the value of any of its indexes). [MH][BR]
  • Ability for user to see the "description" field (or other fields as the developer selects) when inputting data into an edit table. This would provide the user an explanation of the data being input without having to go all the way into the variable. --Cwelch 07:15, 20 January 2009 (PST)cwelch@summitblue.com


  • Color coded lists of node inputs and outputs within the object window (i.e. fill the node icon to the left of it's name in the list with the color chosen for that node in it's diagram). This helps when nodes have a customized color scheme. For example, when adding another label to an index, I want to be able to easily see all the edit tables dependent on that index (as output nodes, but distinguished from purely expression nodes) so I can go through and change the tables accordingly. If I choose a color scheme that distinguishes edit table nodes from expression nodes (or any other distinction between nodes that I should choose), I should be able to take advantage of this from the object window summary. --Charris 12:28, 17 January 2012 (AEST) **
  • Re-sizeable function finder window with mouse-wheel scrollable lists of functions. . --Charris 12:40, 17 January 2012 (AEST) **

Requests for features now added

  • Optimizer engines available on Analytica Cloud Player. This would allow clients to use optimization models without them having to buy the rather expensive Power Player with Optimizer. Pierre Richard, richardpr@gmail.com
  • When learning to use Analytica, I often have to go to the manual, go to the index at the end, look at the page number for the command I want to know about, then get to that page in the manual (which is not the same page as that page in Adobe Reader). I would like it if there was a command reference that had links for all the commands to the appropriate manual sections. **
Wynship Hillier :wynship@hotmail.com
    • You can now jump to the wiki page for each function in several ways -- for example, select the name of a system function or variable in a Definition, right-click, and select Wiki help on F from the right-click menu.
  • It is great that there is a drop-down menu of inputs readily accessed when defining a node. However, it is limited only to inputs that have had arrows linked to the node. Many times, however, you know what variable you want to include in a node, but it is not readily accessible in your model (i.e., it might be 4 modules away). This requires a lot of hunting around for the variable identifier before you can include it in the equation. It would be great if there were easier access to other variables (not just the ones that have arrows) when defining a node. It wouldn't necessarily have to be in the same input drop-down menu (which is kind of nice since it limits to # of variables from which to choose). It could perhaps be an adjacent drop-down menu that has a list of all of the variables in the entire model. I very often find that I would love to have this feature, as it would save a lot of time when defining expressions within a node. I recognize there is a "find" option in the "other" section (i.e., the drop-down that allows you to pick whether you want a table, expression, choice, other, etc.), but using it to find a variable doesn't work well, since once you go to a variable and select it, it wipes out anything else you've written in the definition of the node. --Cwelch 06:56, 25 January 2009 (PST) **
    • The Expression Assist feature is a big help here.Fredbrunt (talk)
  • Dynamic Key: A switch for formatting a graph that only displays non-zero index elements in the key. For graphs that show values for only a couple of index elements it is impossible to translate colors when there are 20 plus elements in the key. Could you have a switch that dynamically populates the key to elements with non-zero values? I'm constantly having to create graphics in excel for this reason.[BR] ***
    • By default now, a graph does no show elements in the key that have only NULL values. If you want to not show zero values, you can add a function that converts zero to NULL.
  • I would like to see a function that can do bootstrap estimates of variance for use with a small sample. I can't seem to find a way of doing that in Analytica. *

Pierre Richard richardp@dfo-mpo.gc.ca

  • I'm not sure why an input node for an index brings the user to the model page, but all other inputs just show the input data. I think it's cleaner and less confusing if the input node for an index would just show, for instance, the list of labels rather than taking you to the model page. --Cwelch 07:18, 20 January 2009 (PST) **

In Analytica 4.2, clicking on List button (input node for Index variable) opens the Object view for the Index, where you can view and modify the List directly. Note that if an Index I has an input node, and is defined as a List of Numbers or Labels (text), you can add, edit, remove, or reorder items in Index J in an Edit table for any variable indexed by J -- in Browse mode. [MH]

  • It would be great if you could create a stacked area line graph, rather than just a stacked area bar graph. Bar graphs are a bit too clunky sometimes. Stacked area line graphs are pretty commonly used. Also, it might be nice to be able to create pie charts as well. In general, my only real complaint about Analytica is the limited ability to edit graphs -- often times I'll have to export the data into Excel to create a pretty graph, which I'd obviously rather avoid if at all possible. --Cwelch 07:20, 20 January 2009 (PST)
    • Stacked area line graphs are available. Double click on Graph to open Graph setup dialog. In the Chart type tab, set the "Stacked" checkbox. You may also set the "Area fill" checkbox. [MH]
  • Possibility to save model diagrams as image-files (PNG etc.): It is possible to save graphs as image-files but not views from models. Now we need to take a screenshot first and then convert it into image-file before we can add it to web-documents. ** --Jvillman 22:39, 27 June 2007 (PDT)
    • This possible now, when diagram window is in focus, select File->export to save an image of the diagram. Fredbrunt (talk)
Comments


You are not allowed to post comments.