CANOPY
======
## Overview
CANOPY (first named for "Computational Algorithm for Neighborhood-Optimized Programs for Youth") is a algorithm intended to assist policy makers with complex decision-making problems. CANOPY was designed to assist city planning processes which feature:
* the need to "optimally" distribute some resource (e.g. dollars of public funding, or service centers) across space;
* a large number of candidate recipients of that resource (e.g. sites to be funded, or candidate locations);
* a complex objective (e.g. one that considers many factors, like multiple characteristics of local populations, both in terms of quantity and quality) that defines the notion of "optimality".
This repository contains documentation of the technical methods that can be used to generate solutions to these complex problems, as well as code--currently still in development--that demonstrates implementations of these methods with realistic--but mock--data.
A presentation with technical details underlying CANOPY can be downloaded [here](https://github.com/nsmader/CANOPY/raw/master/write-ups/presentation/CANOPY%20Project%20-%20Overview%20Presentation.pptx).
## The Problem
When planning locations and offerings across regions and neighborhoods, major for‐profit retailers maximize profits and marketshare with more than rules of thumb. Their analytics incorporate knowledge about where people live, which are likely customers, and what their tastes are. By contrast, local governments and foundations may well be aware of providers and populations within a given human services domain like Head Start, youth anti‐violence programs, or job training, but lack information about how repositioning existing or new resources can more effectively reach larger, and higher priority populations. Without this information, planners are at risk of ineffective use of resources, overloading resources in some areas, overlooking key populations in others, and generally mismatching type of services to needs of local populations. These are deeper issues than maximizing revenue or brand; child development, public safety, and economic opportunity depend on efficient and effective allocation of these scarce resources.
## Analytical Approach
A comprehensive solution that is practically useful to city‐scale resource planners—which may include both city planners at public agencies, public procurement officials, as well as large grant makers—requires a sophisticated understanding of decision‐making by diverse populations targeted by a given human service; a straightforward, concrete means to predict how well a new configuration of resources would meet their goals; and a tool to help identify what types of configurations obtain the best results. CANOPY meets these needs by integrating three interrelated human and technical domains:
### Crystallization of planning objectives and operating constraints
Both planners and researchers meet regularly to concretely articulate considerations for service provision, including:
* **Goals** - such as the relative value for reaching--and differentially prioritizing--different populations, areas of the city, and engaging different types of services and providers; and
* **Constraints** - reflecting both practical considerations such as the planner's overall budget, maximum enrollment capacity of each provider, and political ones such as identifying proposed resource allocations that do not depart too drastically from the status quo in any given year.
### Analysis of service uptake
An analysis of demand for services identifies which families are likely to take up services, and what choices they would be likely to make if they did. Well‐developed econometric methods analyzing "discrete choice" outcomes make use of historical service enrollment patterns to identify how the probability of service uptake is related to all available data on individual/household circumstances, characteristics of service providers, and features of choice context (e.g. neighborhood safety, transportation networks, etc) across city neighborhoods.
### Optimization algorithms
Optimization algorithms provide recommendations for resource allocations that best meet the planner's goals and constraints, given an understanding of how their population of interest would likely take up those services. This involves a range of methods from the field of operations research, which make use of high performance computing implementations that are customized to the planner's specific decision-making structure, which may include decisions of where--and what type of--new programs should be developed, the amount of resources to position at each existing providers, or enhancements to existing offerings to enhance their draw and/or accessibility.
## Anticipated Outcomes
CANOPY's solutions are delivered with an interactive dashboard that allows planners to compare allocation recommendations and anticipated results based on various goal scenarios. Because every technical solution related to social policy requires human judgment, this dashboard is designed to provide a richly‐informed starting point, from which planners can add more nuanced considerations as they make final deliberations.
As a tool for decision-making, CANOPY expects to reduce slack in use of resources, and improve the ability to reach key constituent populations, even with no more than existing resources. In particular, CANOPY will most likely improve access for "donut hole" populations who can benefit from services, but who get overlooked in favor of disproportionate resource allocation to the most immediately recognized geographies of need.
没有合适的资源?快使用搜索试试~ 我知道了~
用于在邻里设施之间分配资源以针对当地人口的模拟退火方法_R语言_代码_下载
共102个文件
png:31个
pdf:14个
r:7个
1.该资源内容由用户上传,如若侵权请联系客服进行举报
2.虚拟产品一经售出概不退款(资源遇到问题,请及时私信上传者)
2.虚拟产品一经售出概不退款(资源遇到问题,请及时私信上传者)
版权申诉
0 下载量 15 浏览量
2022-06-21
19:16:52
上传
评论
收藏 32.04MB ZIP 举报
温馨提示
CANOPY(最初以“青年社区优化计划的计算算法”命名)是一种旨在帮助政策制定者解决复杂决策问题的算法。CANOPY 旨在协助城市规划流程,其特点是: 需要在空间中“优化”分配一些资源(例如公共资金或服务中心); 该资源的大量候选接受者(例如要资助的地点或候选地点); 定义“最优性”概念的复杂目标(例如,在数量和质量方面考虑多种因素,例如当地人口的多种特征)。
资源推荐
资源详情
资源评论
收起资源包目录
用于在邻里设施之间分配资源以针对当地人口的模拟退火方法_R语言_代码_下载
(102个子文件)
canopy-solution -- bball, full data, 1000 iters 652B
Violent-crimes-per-capita-by-tract.csv 27KB
ball-courts.csv 10KB
CensusTractsTIGER2010_cook.dbf 138KB
cents.dbf 131KB
Chicago_CensusTracts_2010_Centroids.dbf 115KB
sids.dbf 105KB
CommArea_Centroids.dbf 19KB
CommAreas.dbf 16KB
Technical Notes for CANOPY.docx 78KB
CH - Simulated Annealing Memo - CANOPY.docx 63KB
CANOPY Project - 2018-04-17 - Notation Guide.docx 60KB
Technical Notes for CANOPY method.docx 79KB
.gitignore 45B
README.md 6KB
Abraham (2003, thesis) - Algorithmics of Two-Sided Matching Problems.pdf 602KB
Lecture 9 mapping.pdf 491KB
Hemmecke Koppe Weismantel (2009) - Nonlinear Integer Programming.pdf 433KB
technical-notes-on-inference-with-capacity-constraints.pdf 332KB
Introduction to CANOPY - Chicago Allocation to Neighborhood-Oriented Programs for Youth Algorithm - With Maps.pdf 311KB
Introduction to CANOPY - Chicago Allocation to Neighborhood-Oriented Programs for Youth Algorithm.pdf 190KB
ConstructiveHeuristicsForTheTSP.pdf 149KB
CANOPY Allocation Maps.pdf 140KB
CANOPY Allocation Maps.pdf 140KB
Bussieck Pruessner (2003) - Mixed-Integer Nonlinear Programming.pdf 119KB
Maps_R_Final.pdf 13KB
Hunneman (2010, thesis) - Advances in Methods to Support Store Location and Design Decisions.pdf 4.68MB
ACS_2008-2012_SF_Tech_Doc.pdf 3.21MB
Soma et al (2014, JMLR) - Optimal Budget Allocation - Theoretical Guarantee and Efficient Algorithm.pdf 290KB
Chicago Map -OSM.png 1.21MB
wireframe.png 414KB
Allocation table.png 69KB
benchmark.png 25KB
sim button.png 13KB
Allocation_Map_45.png 11KB
Allocation_Map_35.png 10KB
Allocation_Map_15.png 10KB
Allocation_Map_40.png 10KB
Allocation_Map_25.png 10KB
Allocation_Map_30.png 10KB
Allocation_Map_35.png 10KB
Allocation_Map_20.png 10KB
Allocation_Map_5.png 10KB
Allocation_Map_40.png 10KB
Allocation_Map_25.png 10KB
Allocation_Map_20.png 10KB
Allocation_Map_30.png 10KB
Allocation_Map_10.png 10KB
Allocation_Map_45.png 10KB
Allocation_Map_15.png 10KB
Allocation_Map_10.png 10KB
Allocation_Map_50.png 10KB
Allocation_Map_50.png 10KB
Allocation_Map_5.png 10KB
Allocation_Map_0.png 9KB
Allocation comparison chart.png 9KB
Allocation_Map_0.png 9KB
slider.png 6KB
progress bar.png 4KB
progress bar - 82pct.png 3KB
wireframe.pptx 1.18MB
CANOPY Project - Overview Presentation.pptx 1.05MB
CANOPY Project - 2018-04-17 - Spatial Data Science Study Group.pptx 1.03MB
~$wireframe.pptx 165B
CommAreas.prj 497B
Chicago_CensusTracts_2010_Centroids.prj 497B
CensusTractsTIGER2010_cook.prj 479B
CensusTractsTIGER2010_cook.qpj 703B
data-prep-1--source-and-combine-data.R 13KB
declare-canopy-method.R 6KB
set-canopy-methods.R 6KB
run-basketball-demo.R 6KB
run-basketball-demo.r 6KB
run-community-area-demo.R 4KB
generate benchmark graph mock-up for wireframe.R 1KB
youth-and-court-data.Rda 12.61MB
Simulated Annealing Output.Rda 793B
.Rhistory 27KB
.Rhistory 18KB
vignette-of-choice-analysis_basic-to-full.Rmd 12KB
data-prep-2--generate-demand-data.Rmd 11KB
technical-notes-on-inference-with-capacity-constraints.Rmd 7KB
Chicago_CensusTracts_2010_Centroids.sbn 8KB
CommAreas.sbn 900B
Chicago_CensusTracts_2010_Centroids.sbx 572B
CommAreas.sbx 188B
CensusTractsTIGER2010_cook.shp 3.56MB
sids.shp 1.53MB
CommAreas.shp 827KB
cents.shp 22KB
Chicago_CensusTracts_2010_Centroids.shp 22KB
CensusTractsTIGER2010_cook.shx 10KB
Chicago_CensusTracts_2010_Centroids.shx 6KB
sids.shx 6KB
cents.shx 6KB
CommAreas.shx 716B
Simulated Annealing Output 7KB
thoughts on wireframe.txt 906B
recommended-allocation.xlsx 14KB
writeframe-benchmark.xlsx 13KB
共 102 条
- 1
- 2
资源评论
快撑死的鱼
- 粉丝: 1w+
- 资源: 9154
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功