onspres2022.Rmd

---
title: "New Mobility Datasets"
subtitle: "ONS Expert Panel"
author: "Robin Lovelace, University of Leeds <br>[Data Science Fellow at N. 10](https://www.ukri.org/opportunity/esrc-adr-uk-no-10-data-science-fellowships-2021/)"
date: "2022-02-07"
output:
  xaringan::moon_reader:
    nature:
      highlightStyle: github
      highlightLines: true
      countIncrementalSlides: false
---

```{r, include=FALSE}
library(tidyverse)
library(tmap)
tmap_mode("view")
knitr::opts_chunk$set(echo = FALSE)
```

## Bio

I'm an Associate Professor of **Transport Data Science** at the Leeds Institute for Transport Studies (ITS). I have experience not only of data science research but of developing new **methods**, **tools** and **software** to inform policy. 
--

## Application

Becoming an ESRC-ADR UK No.10 Data Science Fellow will allow me to make the best use of my **expertise to support the needs of central government**, especially in relation to ... decarbonisation, active travel and the government's leveling-up agenda.
The government has set-out these priorities ...
the [*Decarbonising transport: setting the challenge*](https://www.gov.uk/government/publications/creating-the-transport-decarbonisation-plan) and the *Levelling Up plan*.

--

Translating the laudable objectives of these documents into deliverable plans requires an evidence-based approach, to achieve maximum impact and value for money **at the local level**, an area in which I have substantial experience.

---

# Background, collaborations, interests

- BSc in Geography (Bristol) and MSc in Environmental Science (York), PhD in energy costs of commuting [2014](http://etheses.whiterose.ac.uk/5027/)
- Limitations of available tools for modelling led to the development of new methodologies and the book Spatial Microsimulation with R ([2016](https://spatial-microsim-book.robinlovelace.net/))
- Co-authored Efficient R Programming ([2016](https://csgillespie.github.io/efficientR/))
- Lead Developer of Propensity to Cycle Tool (PCT) ([2017](https://www.jtlu.org/index.php/jtlu/article/view/862))
- Author of Geocomputation with R ([2019](http://geocompr.robinlovelace.net/transport.html))
- Substantial [research](https://www.robinlovelace.net/publication/) output ([h-index](https://scholar.google.com/citations?user=xDJHVCAAAAAJ&hl=en): 20)

--

```{r, out.width="30%", fig.show='hold', echo=FALSE}
knitr::include_graphics(c(
  "https://docs.ropensci.org/stplanr/reference/figures/stplanr.png",
  "https://raw.githubusercontent.com/ropensci/stats19/master/man/figures/logo.png",
  "https://github.com/Robinlovelace/geocompr/blob/main/images/geocompr_hex.png?raw=true"
))
```

--

- Evidence-based policies in government: [Data Science Fellowship at N. 10](https://www.ukri.org/opportunity/esrc-adr-uk-no-10-data-science-fellowships-2021/)

???

- Turing Fellowship 
- LIDA internship on open transport infrastructure data
- Links with DfT, MHCLG, TfNH, international partners

--

- Future areas of development: Reproducible Bayesian modelling of proportions (Dirichlet regression), Machine Learning, Decarbonisation Agenda

---

---

.left-column[

### Expertise: Geocomputation

Computationally efficient data science tools

Intuitive, future proof, scalable code

Geographic vector data analysis

![](https://user-images.githubusercontent.com/1825120/141863981-31a11068-a5f3-4e19-9471-b03dbb36f9a8.png)
Geographic raster data

![](https://geocompr.robinlovelace.net/02-spatial-data_files/figure-html/raster-intro-plot-1.png)

]

--

.right-column[

```{r}
knitr::include_graphics("https://user-images.githubusercontent.com/1825120/142071229-81358e26-5e8d-437e-9ef8-91704a4e690f.png")
```

Source: Morgan and Lovelace (2020) https://doi.org/10.1177/2399808320942779

]

???

I am an all-round data scientist with experience with Python, Julia, and command line tools such as Docker and shell scripting for scalable data science applications.

I have particular expertise in R and geocomputation with R in particular.

---

### Experience working with OD data

Adding value and detail to existing OD data.
Source: Lovelace, Félix and Carlino ([2022](https://osf.io/qux6g/) preprint)

![](https://github.com/Robinlovelace/odjitter/raw/responses/README_files/figure-gfm/jittered514-1.png)

---

### From OD data to travel behaviour

Source: Lovelace, Félix and Carlino ([2022](https://osf.io/qux6g/) preprint)

![](https://github.com/Robinlovelace/odjitter/raw/responses/README_files/figure-gfm/rneted-1.png)

---

### Validating synthetic OD datasets

Source: Lovelace, Félix and Carlino ([2022](https://osf.io/qux6g/) preprint)

![](https://github.com/Robinlovelace/odnet/raw/main/README_files/figure-gfm/rnets-1.png)

---

### Scalability vs resolution

Source: [UKRI CREDS project repo](https://github.com/creds2/od-data)

![](https://github.com/creds2/od-data/raw/master/overview_map1.png)

---

### OA-WPZ data

There are 17,848,366 OA to WPZ records, 170k OAs, 54k WPZ

For 5km buffer around London, 1.5 million OD pairs with destinations

[![](https://user-images.githubusercontent.com/1825120/152778656-603196b0-26d3-4c49-8940-6c0d3c9e1ce5.png)](https://rpubs.com/RobinLovelace/863109)

---

### Reproducible example

.left-column[


```{r, echo=TRUE}
u = "https://github.com/ITSLeeds/od/releases/download/v0.3.1/od_intra_top_sf.geojson"
desire_lines_oa_wpz_1k = sf::read_sf(u)
oas_in_buffer = sf::read_sf("https://github.com/ITSLeeds/od/releases/download/v0.3.1/oas_in_buffer.geojson")
wpz_in_buffer = sf::read_sf("https://github.com/ITSLeeds/od/releases/download/v0.3.1/wpz_in_buffer.geojson")
library(tmap)
tmap_mode("view")
m = tm_shape(desire_lines_oa_wpz_1k) +
  tm_lines() +
  tm_shape(oas_in_buffer) + tm_dots(col = "darkgreen") +
  tm_shape(wpz_in_buffer) + tm_dots(col = "darkred")
```

]

.right-column[

```{r}
m
```

]


---

### Data sources for movement estimates

.pull-left[

## Open data

- Traffic count data
- Urban Observatory type data (Newcastle, Birmingham, Manchester)
- Faceboook and Google open mobility data
- 'OSM2od' - spatial interaction model
- Modelled data
  - jittering: spatial disaggregation
  - temporal disaggregation
- Scenarios of change - e.g. Go Dutch open data estimates of cycling potential

]

.pull-right[

## Proprietary data

- National Travel Survey
- Mobile Telephone Data
- Large GPS type data (biobank, Google timeline, Straval)

]

---

## From OD data to policy tools

.left-column[

Lead Developer of the DfT's PCT (Lovelace et al. 2017) : transformational impact on planning in the UK (source: REF Impact Case Study)

COVID response: RAPID tool (Lovelace et al. 2020)

ActDev tool for informing planning process

]

.right-column[

<iframe width="560" height="315" src="https://www.youtube.com/embed/nNYroA16JEQ?start=120" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>

]

---

# Open source software, community building

- New high performance libraries

![](https://user-images.githubusercontent.com/1825120/142075245-3dee4b6e-e4d6-424d-b4aa-a72458e88fd6.png)

---

.left-column[
## OD data and policy

Close working relationship with DfT over past 6 years

Collaborations with Ordnance Survey (LIDA), National Highways (training), links with DLUHC and Open Innovation Team

Transport behaviour, infrastructure, safety, energy

]
.right-column[

![](https://raw.githubusercontent.com/cyipt/actdev/main/data-small/scenario-overview.png)

![](https://user-images.githubusercontent.com/1825120/142076842-bc87bf5b-f8b2-4fbc-bcb4-a4da575b171f.png)

]

---

.left-column[

# Ideas

Hackathons

Active travel

Road safety policy

Post COVID recovery

Levelling up metrics

Nature recovery networks

Citizen science and data literacy

]

.right-column[

![](https://saferactive.github.io/trafficalmr/articles/covid.png)

![](https://a-b-street.github.io/docs/project/history/retrospective/edit_roads.gif)

]

---

## Example: Climate change, what data science can bring

.left-column[

Emotive and potentially polarising issue

Data can provide a shared starting point

Nationally scenarios are vital

Locally, visions, trust, buy-in and participation are key

]

.right-column[


<!-- <iframe src="https://ourworldindata.org/grapher/annual-co-emissions-by-region" loading="lazy" style="width: 100%; height: 600px; border: 0px none;"></iframe> -->

```{r, eval=FALSE, echo=FALSE}
piggyback::pb_upload("annual-co-emissions-by-region.csv")
piggyback::pb_download_url("annual-co-emissions-by-region.csv")
```

```{r, include=FALSE}
emissions = readr::read_csv("https://github.com/Robinlovelace/presentations/releases/download/2020-02/annual-co-emissions-by-region.csv")
emissions = emissions %>% rename(Emissions = `Annual CO2 emissions (zero filled)`)
emissions_regions = emissions %>%
  filter(is.na(Code)) %>%
  filter(Entity != "World") %>% 
  filter(!str_detect(string = Entity, pattern = "EU|Fr|Ku|Leew|exc|Pan|Ry|St"))
```

```{r}
g = ggplot(emissions_regions) +
  geom_line(aes(Year, Emissions, colour = Entity), show.legend = FALSE)
plotly::ggplotly(g)
```

]

???

The Downward trend during coronavirus saw emissions drop by ~10%

The scale of the challenge is to reduce emissions by 10% every year, every year, and the challenge gets hard with each year

That means: we need transformational

---

### Decarbonisation

<iframe src="https://ourworldindata.org/grapher/co2-mitigation-2c" loading="lazy" style="width: 100%; height: 600px; border: 0px none;"></iframe>

---

# Making UK data come to life

![](https://github.com/zonebuilders/zonebuilder/releases/download/v0.0.2.9000/cityscale.png)
Source: Lovelace, Tennekes, Carlino ([under review](https://zonebuilders.github.io/zonebuilder/articles/paper.html))

![](https://user-images.githubusercontent.com/1825120/141021386-3ee6af6d-8ca8-4f4c-9856-c636f4a8fc0e.png)


---
class: center, middle

# Thanks, look forward to working with you  🖧 + 📈 + ✨ =  🚀!

--

## References

--

Lovelace, R., Goodman, A., Aldred, R., Berkoff, N., Abbas, A., Woodcock, J., 2017. The Propensity to Cycle Tool: An open source online system for sustainable transport planning. Journal of Transport and Land Use 10. https://doi.org/10.5198/jtlu.2016.862


--

Morgan, M., Lovelace, R., 2020. Travel flow aggregation: nationally scalable methods for interactive and online visualisation of transport behaviour at the road network level. Environment & Planning B: Planning & Design. https://doi.org/10.1177/2399808320942779

--

Lovelace, R., Tennekes, M., Carlino, D., 2021. ClockBoard: a zoning system for urban analysis. https://doi.org/10.31219/osf.io/vncgw

--

Lovelace, Robin, Rosa Félix, and Dustin Carlino. “Jittering: A Computationally Efficient Method for Generating Realistic Route Networks from Origin-Destination Data.” OSF Preprints, January 13, 2022. https://doi.org/10.31219/osf.io/qux6g.