From @Ennazus on September 24, 2014 18:0

assign them in the same way

Copied from original issue: #173

From @Ennazus on December 4, 2014 17:26

Copied from original issue: #214

From @Ennazus on August 27, 2014 0:13

Waiting for Hana to provide the inputs

Copied from original issue: #160

From @Ennazus on November 7, 2014 16:17

Copied from original issue: #192

From @Ennazus on November 7, 2014 21:24

Copied from original issue: #195

From @Ennazus on July 24, 2014 22:31

Copied from original issue: #141

From @Ennazus on November 14, 2014 17:1

record run time and substantive differences.
It may be that we never need to run 100% populations

Copied from original issue: #197

From @bricegnichols on November 13, 2014 1:19

Might be useful to have total daily model results in a single project. We could generate our overall maps and worksheets from here and optionally provide detailed TOD results as needed.

Copied from original issue: #196

From @bricegnichols on October 8, 2014 15:13

This has killed some of the test runs. We should try to catch this early on.

Copied from original issue: #177

From @Ennazus on November 20, 2014 23:45

Copied from original issue: #203

From @bricegnichols on July 22, 2014 0:11

Copied from original issue: #138

From @billyc on November 26, 2014 18:31

All open source code needs a license and copyright attribution. We should:

1. Decide on a license. I suggest Apache
2. Grab generic copyright boilerplate and assign all copyright to PSRC.
3. Paste that copyright notice into every single source file.

Copied from original issue: #208

From @Ennazus on November 25, 2014 23:5

Should we merge these repositories so we can pull/merge more easily?

Copied from original issue: #207

From @Ennazus on May 29, 2014 18:57

Copied from original issue: #116

From @Ennazus on November 24, 2014 17:22

We need to set up some code so that we can easily debug a partially complete crashed run

Copied from original issue: #205

From @Ennazus on December 5, 2014 16:32

it's very hard-coded
very difficult to change
and hard to understand

Copied from original issue: #215

From @stefancoe on July 18, 2014 19:35

How to run a scenario from top to bottom.

Copied from original issue: #135

From @Ennazus on December 2, 2014 23:37

This is so we can run the Transportation Futures Project using SoundCast in shadow mode

Copied from original issue: #211

From @Ennazus on September 23, 2014 17:31

We could take a look at the NPMRDS HERE data, which is link level travel time data. Gary Simonson approached us a while back about how we might use this data and I ended up spending a little time on it. You start with a month’s worth of data and then whittle it down to the days/time periods of interest and calculate link level speeds. Of course the network does not match ours, but it would be pretty easy to create a crosswalk for a subset of the data. It seems like good (and free) speed data is hard to come by and could be a nice complement to counts.

-Stefan

The documentation here is terrible and I think Gary has some better stuff. I have already pinged him….Also, it’s not just freight datat as the link suggests.

http://www.ops.fhwa.dot.gov/freight/freight_analysis/perform_meas/vpds/npmrdsfaqs.htm

Copied from original issue: #171

From @stefancoe on December 8, 2014 16:52

Copied from original issue: #217

From @Ennazus on August 6, 2014 19:16

where models are run via a web-type interface

Copied from original issue: #154

From @bricegnichols on July 22, 2014 18:51

Copied from original issue: #139

From @Ennazus on October 2, 2014 17:33

Short term solution: factor enlisted jobs;
Longer term: Remove military personnel from synthetic population; maybe make JBLM an external and distribute trips based on data;
Look into what San Diego and Other places with ABMs and large military bases

Copied from original issue: #175

From @Ennazus on October 16, 2014 17:39

Copied from original issue: #182

From @bricegnichols on November 7, 2014 18:4

I moved a couple hard codes from the generation script that never made their way into the final input_configuration script. This should be all we need:

Special generator zones and demand (dictionary key is TAZ, value is demand)

spg_general = {3110: 1682,
631: 7567,
438: 14013}
spg_airport = {983: 101838}

Copied from original issue: #193

From @Ennazus on November 7, 2014 16:17

Copied from original issue: #192

From @Ennazus on November 14, 2014 17:1

record run time and substantive differences.
It may be that we never need to run 100% populations

Copied from original issue: #197

From @Ennazus on November 25, 2014 22:53

make all places in code consistent and the smallest number of places

Copied from original issue: #206

From @Ennazus on September 23, 2014 18:0

so only include the jobs that come from workers who don't live on base

Copied from original issue: #172

From @Ennazus on December 2, 2014 23:40

We need to prepare 2014 inputs in this case.

Copied from original issue: #212

From @Ennazus on August 6, 2014 18:48

(Get Joe's help too.)

Copied from original issue: #151

From @Ennazus on November 20, 2014 23:31

Copied from original issue: #202

From @Ennazus on December 2, 2014 23:41

Do we want to consider any re-design of models?

Copied from original issue: #213

From @Ennazus on July 24, 2014 22:29

Copied from original issue: #140

From @bricegnichols on November 7, 2014 19:44

Currently, supplemental trips depend on a set of local skims. The only time this is an issue is if we create a new repo and run supplemental trips only (without first doing some skimming with seed trips). We need to either try/catch this problem or redirect to use seed skims.

Copied from original issue: #194

From @Ennazus on November 20, 2014 23:50

Copied from original issue: #204

From @Ennazus on July 24, 2014 22:32

• Community Transit
• Stops
• Everett Transit

Copied from original issue: #142

From @Ennazus on November 3, 2014 16:13

Our transportation friends are very interested in a measure like this.
It will be fairly easy to produce, but maybe not theoretically strong.

Copied from original issue: #189

From @Ennazus on August 6, 2014 19:17

Copied from original issue: #155

From @stefancoe on November 18, 2014 18:42

These scripts should be using the stand alone EmmeProject class.

Copied from original issue: #199

From @Ennazus on June 11, 2014 17:58

Do a run with and without tolls on:

1. Tacoma Narrows
2. 520

Observe- change in toll volumes on both links
Compare to actual before -after's (Kris O says he can get the data)

Copied from original issue: #120

From @Ennazus on November 14, 2014 17:14

Copied from original issue: #198

From @Ennazus on July 31, 2014 22:10

see the run on ModelSrv1\soundcat_new

Copied from original issue: #146

From @Ennazus on July 17, 2014 15:25

so we don't need the banks as much

maybe Brice could do this too.
thanks!

Copied from original issue: #133

From @Ennazus on September 16, 2014 18:11

o How are we going to manage this module extension for (1) future-year scenarios and (2) base year re-estimation? I think future-year scenarios are easy enough to handle if we assume something simple like growth rates, which is how externals are usually handled. Maybe this can be part of a more complex visitor model that we eventually estimate.

Copied from original issue: #166

From @Ennazus on August 6, 2014 18:49

Copied from original issue: #153

From @Ennazus on November 20, 2014 23:20

Copied from original issue: #200

From @stefancoe on November 26, 2014 22:27

Add this to network_importer.py

Copied from original issue: #210

From @bricegnichols on November 6, 2014 17:42

Input processing is too convoluted, want to streamline this better for future-year runs

Copied from original issue: #190

From @Ennazus on November 20, 2014 23:31

make a few example charts
investigate
comparison to base

Copied from original issue: #201