1/15/2010 11:03:51 AM
Pete Jackson
Posts: 7
|
I need to look at access to food shops and GP surgeries. For a realistic approach it would be useful to look at access to the destination and the ability to return to the origin in a reasonable amount of time e.g. get to the shops in the morning and return within at the most a couple of hours. I don't think there is, but is there anyway that accession could measure for example whether somebody could make a return journey to and from a destination within a set period of say two hours?
|
|
permalink
|
1/15/2010 11:14:25 AM
keithdrew76
Posts: 44
|
Pete
In an urban network I doubt it matters so much as you can pretty much assume a network is two way (I think). I guess in an rural area its a little more tricky with infrequent services. Alkso tricky if you looking at individual destinations (i.e. can you get to Spar on the High Street there and back).
It sounds like you need to do an 'origin-destination' calc say 10 to 11am and then reverse it 11 to 12pm (you can do this in the calc wizard by detrmining the direction of the calculation), and then run full reports of both and compare where the Origin and Destination match, and if they pass the test that you set (ie can get there in x minutes and can get back in x minutes).
Hope that helps.
Keith
|
|
permalink
|
1/15/2010 11:39:35 AM
Pete Jackson
Posts: 7
|
Keith, Thanks very much, I'm looking at a rural network for Dorset and I think that I would have to look at your second answer.
Also in relation to Frequency Networks how do I get to use these - we import data as an Atco Cif file does this support Frequency networks or do I have to use ACC Map and if so how do I get hold of data in this format, thanks
Pete
|
|
permalink
|
1/15/2010 11:56:42 AM
keithdrew76
Posts: 44
|
Hi Pete
By frequency networks do you mean catchments around stops that serve a particular destination (direct bus routes buffered say 800metres)? If thats the case I tend to import the bus network to Accession, but then 'break' into accession and export out the bus network - I then do the analysis in MapInfo. But you can create these in Accession through the local calculation where you set your threshold and time periods.
I personally have never used ACC map so not sure what data is required - but sure others do on here :-)
Cheers
Keith
|
|
permalink
|
1/15/2010 12:08:18 PM
Pete Jackson
Posts: 7
|
Keith,
No what I was referring to was that when you undertake a calculation in Accession you can undertake it using a timetabled based network or a frequency based network. I've always used a timetabled based network but to use a Frequency based network you have to import I think from ACMAP data rather than ActoCIF. I was just confused as to how I should go about this as I don't even know how to get data i this format - don't worry too much.
Thanks
Pete
|
|
permalink
|
1/15/2010 12:14:31 PM
keithdrew76
Posts: 44
|
No worries. I could be completey wrong but think the frequency option is for those areas that don't have detailed data (timetable) which I think was an issue in the early days of Accession for some areas. I have ever only used the timetable option.
I think the solution to your query is to do the two sets of calcs (O-D, D-O) and then do the main report (if you can I'd recommend going through an Access Query to generate your main reports to avoid any issue with multiple tabs in Excel) to do the lookups...unless Accession not lets you generate .csv/.txt reports.
Good look with the study!
Keith
|
|
permalink
|
1/15/2010 1:12:43 PM
admin
Administrator
Posts: 146
|
Hi,
To add clarification the the Frequency network is not reallyused any more, Accmap does create a frequency based network from a CIFfile. Though the problem with Frequency networks is that on a lowfrequency service such as 1PH this will give a large wait time. Using atimetabled network is a much more accurate way of producing anaccessibility assessment.
For the checking the main tablereports to get the same OD - DO result there is a council which hasperfected this technique and written an application to do it, i thinkit is Norfolk though am not certain of this. Hopefully they will getinvolved in this discussion.
Regards,
Dan
|
|
permalink
|
3/3/2010 4:15:03 PM
Jonno1
Posts: 14
|
What you could do is to do 4 or more Accession runs, 2 say 0700-0900 and 1000-1200 for outward journeys, 2 say 1300-1500 and 1600-1800 for return journeys. Then export the relevant results table into an Access database, and perform queries on the data to identify those OD pairs which have a minimum of 1 outward and 1 return journey ie no '100000000' in the data. Then take the mean/median? of these values and then use these in the standard threshold calculations done using MS Access?
|
|
permalink
|
3/3/2010 4:21:47 PM
Jonno1
Posts: 14
|
admin wrote:
Hi,
'For the checking the main tablereports to get the same OD - DO result there is a council which hasperfected this technique and written an application to do it, i thinkit is Norfolk though am not certain of this. Hopefully they will getinvolved in this discussion'
What I would do would be to export the results into an Access database, and then rearrange the order of the columns in the results by writing a query in the design grid. To create OD pairs I would write an expression in the query design grid along the lines of 'OriginID' & '_' & 'DestinationID' which will give results in the format '451_900'. You should then be able to compare results for each OD pair for different time periods.
|
|
permalink
|