The "Alpaca" project, funded by the Lincoln Institute of Land Policy (http://www.lincolninst.edu/), developed software designed to facilitate the integration of bid-rent land use models with scenario planning tools. Such models can enhance scenario planning efforts by simulating how real estate markets work, adding a variety of metrics relating to topics such as:
- Property value and rent
- Socio-economic characteristics of tenants
- Housing affordability and cost burden
This repository provides a homepage for the code developed in the course of the project, including:
- mu-land: the "core" bid-rent model evaluation software
- muLandWeb: a Python wrapper for mu-land allowing models to be offered as web services callable by third-party tools
- example: client-side Python scripts developed in case studies, showcasing various methods of using mu-land
The server-side software (mu-land and muLandWeb) has been tested using Linux, specifically Ubuntu 14.04 LTS. Other distributions and/or operating systems may work, but there are no guarantees. Some of the client-side scripts demonstrate how models could be called from other environments.
The mu-land core uses Boost 1.54 libraries.
PostgreSQL is used for back-end database management and spatial operations involving PostGIS.
A bid-rent land use model is required in order to apply the framework. Regional planning agencies using the Cube Land software distributed by Citilabs (see http://www.citilabs.com/) may have existing models in a format very similar to that required by mu-land. Manhan Group, LLC (http://manhangroup.com/) has developed many of these models and provides related consulting services.
The mu-land program must be installed first. Follow the instructions in the "readme" file within the mu-land folder.
Then, do the following:
- Edit /etc/apt/sources.list, adding
deb http://apt.postgresql.org/pub/repos/apt/ trusty-pgdg main
- Install required packages:
wget --quiet -O - https://www.postgresql.org/media/keys/ACCC4CF8.asc | apt-key add -
apt-get update
apt-get dist-upgrade
apt-get install python3 python3-dev python3-pip python-virtualenv postgresql-9.5 postgresql-9.5-postgis libpq-dev
- Configure PostGIS as needed.
- Install muLandWeb:
mkdir mulandweb
cd mulandweb/
virtualenv -p `which python3` env
. env/bin/activate
pip install https://github.com/ManhanGroup/Alpaca/archive/master.zip
- Create mulandweb/bin directory and copy mu-land into it as muland
- Initialize tables at the database using
python -m mulandweb -c
- Unpackage model into mulandweb directory and import e.g.
python -m mulandweb -i fresno --import-srid 900913
- Software design/testing:
- Colby Brown (colby@manhangroup.com)
- Pedro Donoso (pedrodonosos@gmail.com)
- Programming:
- Felipe Saavedra (fsaavedr@dcc.uchile.cl)
- Leandro Lima (leandro.lima@toptal.com)