Generate Spatially Efficient Experimental Designs • speed

Overview
Installation
Features
Examples
How It Works
Citation
License

Overview

The speed package optimises spatial experimental designs by rearranging treatments to improve statistical efficiency while maintaining statistical validity. It uses simulated annealing to:

Minimise treatment adjacency (reducing neighbour effects)
Maintain spatial balance across rows and columns
Respect blocking constraints if specified
Provide visualisation tools for design evaluation

Installation

You can install the development version of speed from GitHub with:

# pak
if (!require("pak", quietly = TRUE)) install.packages("pak")
pak::pak("biometryhub/speed")

# devtools
if (!require("devtools", quietly = TRUE)) install.packages("devtools")
devtools::install_github("biometryhub/speed")

# remotes
if (!require("remotes", quietly = TRUE)) install.packages("remotes")
remotes::install_github("biometryhub/speed")

Features

Flexible optimisation of experimental designs
Support for blocked designs
Customisable optimisation parameters
Built-in visualisation functions
Progress tracking during optimisation
Early stopping when convergence is reached

See the package documentation for more detailed examples and options.

Examples

Basic

A simple example optimising a 4×3 completely randomised design with 4 treatments:

library(speed)

# Create a simple design with 3 replicates of 4 treatments
df <- data.frame(
  row = rep(1:4, each = 3),
  col = rep(1:3, times = 4),
  treatment = rep(LETTERS[1:4], each = 3)
)

# Optimise the design with seed for reproducibility
result <- speed(df, "treatment", seed = 42)
#> row and col are used as row and column, respectively.
#> Optimising level: single treatment within whole design 
#> Level: single treatment within whole design Iteration: 1000 Score: 1 Best: 1 Since Improvement: 475 
#> Level: single treatment within whole design Iteration: 2000 Score: 1 Best: 1 Since Improvement: 1475 
#> Early stopping at iteration 2525 for level single treatment within whole design

# Plot the optimised design
autoplot(result)


# View optimisation progress
plot_progress(result)

Blocked design

You can also optimise designs within blocks:

# Create a design with blocks
df <- data.frame(
  row = rep(1:6, each = 4),
  col = rep(1:4, times = 6),
  treatment = rep(LETTERS[1:8], 3),
  block = rep(1:3, each = 8)
)

# Optimise while respecting blocks
result <- speed(df,
  "treatment",
  swap_within = "block",
  iterations = 5000,
  seed = 42
)
#> row and col are used as row and column, respectively.
#> Optimising level: single treatment within block 
#> Level: single treatment within block Iteration: 1000 Score: 2.571429 Best: 2.571429 Since Improvement: 543 
#> Level: single treatment within block Iteration: 2000 Score: 2.571429 Best: 2.571429 Since Improvement: 1543 
#> Early stopping at iteration 2457 for level single treatment within block

# Plot the design with block boundaries
autoplot(result)

More Examples

For more detailed examples, see the getting started vignette or the vignette about more complex examples including, but not limited to:

How It Works

The speed package uses simulated annealing to optimise experimental designs by evaluating and improving layouts based on an objective function. By default the optimisation process:

Evaluates designs using a score (lower is better) that combines:
- Adjacency score: Penalises treatments appearing next to each other (reduces neighbour effects)
- Balance score: Rewards even distribution of treatments across spatial factors (rows, columns, blocks)
Proposes changes by swapping treatments between plots
Accepts improvements and occasionally accepts worse solutions to avoid local optima
Converges to an optimised layout through gradual “cooling” of the acceptance threshold

You can customise the objective function and optimisation parameters to suit specific experimental needs (see vignettes for details).

Citation

If you use speed in your research, please cite:

Warning in citation("speed"): could not determine year for 'speed' from package
DESCRIPTION file
To cite package 'speed' in publications use:

  Rogers S, Taylor J, Edson R, Pipattungsakul W (????). _speed:
  Generate Spatially Efficient Experimental Designs_. R package version
  0.0.3, <https://biometryhub.github.io/speed/>.

A BibTeX entry for LaTeX users is

  @Manual{,
    title = {speed: Generate Spatially Efficient Experimental Designs},
    author = {Sam Rogers and Julian Taylor and Russell Edson and Wasin Pipattungsakul},
    note = {R package version 0.0.3},
    url = {https://biometryhub.github.io/speed/},
  }

License

This project is licensed under the MIT License - see the LICENSE file for details.

Code of Conduct

Please note that the speed project is released with a Contributor Code of Conduct. By contributing to this project, you agree to abide by its terms.

speed - Spatially Efficient Experimental Designs

Table of Contents