topic Re: summarize the attribute values of lines (in one layer) which have the same start- an endpoint in Data Management Questions

summarize the attribute values of lines (in one layer) which have the same start- an endpoint

JohannesUhl1 — Tue, 23 Jun 2015 13:38:02 GMT

Do you know whether it is possible to summarize the attribute values of lines which have the same start- an endpoint?

This is my sitaution: I have ONE layer with lots of buslines between german cities. Now I want ArcGIS to summarize all lines which run between two cities to one line with the summarized frequentation attribut value. An this for all city to city relations.

Example - actual date situation:

Line 1: Stuttgart-München -> 5 times a week

Line 2: Stuttgart-München -> 2 times a week

Line 3: Stutgart-Hamburg -> 3 times a week

Line 4: Stutgart-Hamburg -> 3 times a week

Line 5: Stutgart-Hamburg -> 3 times a week

Result shall be:

one line between Stuttgart and München with attribut value 7 times a week and one line between Stuttgart and Hamburg wit attribut value 9 times a week.

I already tried lots of options, but none worked (including Dissolve).

Thank you very much for ideas,

Johannes

Re: summarize the attribute values of lines (in one layer) which have the same start- an endpoint

SepheFox — Tue, 23 Jun 2015 13:59:05 GMT

It would be easier to answer if I could see a screenshot of your table, with the fields and attributes, but assuming you have fields of start/end, and times per week, you could use the summary statistics tool to obtain the sum of times per week, with the start/end as a case field. If the start and end are different fields, you will need to add a new field and calculate it to have both values in it.

Re: summarize the attribute values of lines (in one layer) which have the same start- an endpoint

JohannesUhl1 — Tue, 23 Jun 2015 14:19:36 GMT

Thank you very much for your anwer.
I guess your idea doesn´t work, because the table doesn´t include any start- an endpoint information. This information is only included in the shapefile, but not shown in the table. In my opinion it´s not possible to work with the data in the table. Thats why I guessed it´s neccessary to work with the spatial data.ot:

To the screenshot:
"Fahrten_Wo" is the number of times per week. All other columms are not used. I already splitted up the entire buslines to lines between single cities. Thats why you see the line "EU 2060 Mannheim-Kosice" 8 times: its sepated by the intermediate stops (Mannheim-Stuttgart (FID 1), Stuttgart-München (FID 2), and so on). But this names of the intermediate stops are not in the table

Re: summarize the attribute values of lines (in one layer) which have the same start- an endpoint

SepheFox — Tue, 23 Jun 2015 14:26:37 GMT

I'm sorry, I don't really understand what you're saying. Your original question didn't mention a table different from your shapefile. I can't really help you in less I'm clear what you are working with.

Re: summarize the attribute values of lines (in one layer) which have the same start- an endpoint

JohannesUhl1 — Tue, 23 Jun 2015 14:44:28 GMT

Sorry, im quiete new in working with GIS.
The table and the shapefile aren´t different. What I just wanted to say is, that there aren´t any position data (like koordinates or citynames) in the table. That´s why I thought your idea can´t work.

Re: summarize the attribute values of lines (in one layer) which have the same start- an endpoint

SepheFox — Tue, 23 Jun 2015 14:50:16 GMT

So what is the field called "Liniename"?

Re: summarize the attribute values of lines (in one layer) which have the same start- an endpoint

JohannesUhl1 — Tue, 23 Jun 2015 14:59:53 GMT

The field called "Linienname" ist just the name of the WHOLE line. It consist oft the linename/number itseslf and the very first city and the very last city. Thats why I can´t use this field: I want to know the sum of all runs a week (sum of several companies) between two cities in Germany. Would I just use the cities named in the field "Linienname" I would ignore the connections between the intermediate stops of the lines.

By the way: Thank you very much for your patience!

Re: summarize the attribute values of lines (in one layer) which have the same start- an endpoint

NICOLAPEDDIS — Tue, 23 Jun 2015 15:02:27 GMT

Sephe, the "Liniename" means-->> "line name". It's like a bus route name

johannesuhl , do you have a common value between the table and the polyline..??? Maybe you can make a joint between the polyline and the external table, then you can try to summarize the routes.

Re: summarize the attribute values of lines (in one layer) which have the same start- an endpoint

RichardFairhurst — Tue, 23 Jun 2015 15:04:17 GMT

I think I kind of understand what you are working with. You split a larger bus route into multiple lines based on city intermediate stops, but did not insert the intermediate stops into the table. Presumably different several bus lines connect the same intermediate destination pairs for at least part of their route so they should have a way to be summarized to get counts from your data.

Assuming the points shown on your exhibit are the intermediate stops you mentioned, you can use the Spatial Join tool to get their attributes on to the lines. Use the Spatial Join Tool with the One To One option and make the Lines the Target features and the Points the Join Features. Any points that do not actually represent stops that you split the lines with should be excluded when the tool is run so that only the points at the start and end of each line are used.

In the Field Map of the Spatial Join tool right click the field to adjust the properties for the City Name field (or whatever field works to identify each intermediate stop) to make it a Text Field with 254 characters and use Join as the Merge Rule Option with a delimiter character, such as a semicolon or a dash. This will concatenate the two stop names together into a new field on the lines. You may want to alter the Name order in the field, since the concatenation won't really be based on any kind of driving directions or alphabetical order, just the order it encountered the points. You then should be able to use those to summarize the intermediate stop segments based on that field.

Re: summarize the attribute values of lines (in one layer) which have the same start- an endpoint

SepheFox — Tue, 23 Jun 2015 15:16:48 GMT

You might try using the summary statistics, but use the shape_length field as the case field. It is unlikely that you have the same length for features that are not coincident.

Re: summarize the attribute values of lines (in one layer) which have the same start- an endpoint

JohannesUhl1 — Tue, 23 Jun 2015 15:31:46 GMT

Thank you very much, Richard! Thats exactly the problem I have!

Your solution sounds great, I just tried it. But could you describe more exactly, who I can alter the name order in the field? One line ist now for example called "Stuttgart;München" an another "München; Stuttgart". Is it possible to make both to "München; Stuttgart"? For example, by putting the names within each cell in an alphabetically order?

Re: summarize the attribute values of lines (in one layer) which have the same start- an endpoint

JohannesUhl1 — Tue, 23 Jun 2015 16:25:53 GMT

Thank you very much for this creative idea. I will try 🙂

Re: summarize the attribute values of lines (in one layer) which have the same start- an endpoint

JohannesUhl1 — Tue, 23 Jun 2015 16:26:50 GMT

No, as far as I know there is no common value between these...

Re: summarize the attribute values of lines (in one layer) which have the same start- an endpoint

RichardFairhurst — Tue, 23 Jun 2015 17:16:56 GMT

First check that no more than 2 names got concatenated, since if you have cases where more than that occur you probably should figure out why manually. For ordering the cities alphabetically; first select all records that contain a semicolon and space in the output (i.e., CITY_NAME Like '%; %') to make sure no records with just a single name are included, then you can do the following Python calculation to reorder the two names alphabetically (make the field name match your field name of course):

sorted(!CITY_NAME!.split('; '))[0] + '; ' + sorted(!CITY_NAME!.split('; '))[1]