What is it about?
This paper explains a way to use prior information and simulated surveys to optimize the way archaeologists (and search-and-rescue planners) assign their limited search resources (like person-days) to find scarce or hard-to-find "targets." In the example used here, the targets are early and mid-Holocene archaeological sites in northern Jordan. The article also introduces two online apps that help to automate this process.
Featured Image
Photo by NEOM on Unsplash
Why is it important?
Archaeologists typically have very limited resources for surveys and, if they are allocated only randomly or evenly, there is high risk that their surveys will miss rare or unobtrusive sites.
Perspectives
I have been working on theoretical aspects of search and survey for a long time, but worry that most archaeologists find the methods that help us conduct surveys more efficiently are too difficult or require too much math. I am therefore very pleased that my colleagues and I have found ways to automate some aspects of this work and I hope that this will make these methods much more accessible.
Edward B Banning
University of Toronto
Read the Original
This page is a summary of: Bayesian optimal allocation of effort in archaeological survey: A case study in Wadi Quṣaybah, Jordan, Journal of Archaeological Science, July 2026, Elsevier,
DOI: 10.1016/j.jas.2026.106568.
You can read the full text:
Resources
SurveyPlanningAllocCostingAndEval/optimal_allocation: v1.0.2 – UI Enhancements and Allocation Description Improvements
The purpose of the application is to provide a complete, reproducible, and user-friendly workflow for executing the Bayesian optimal-allocation method in the field and for updating the predictive model after each iteration of survey. This appendix provides an overview of the application's structure, inputs, outputs, and internal logic.
SurveyPlanningAllocCostingAndEval / sweep_width Public archive
The full R script used to generate the sweep-width estimates reported in the main text.
SurveyPlanningAllocCostingAndEval / optimal_allocation Public archive
The allocation module described in Section 3 is implemented in an R Shiny application developed specifically for this study. The purpose of the application is to provide a complete, reproducible, and user-friendly workflow for executing the Bayesian optimal-allocation method in the field and for updating the predictive model after each iteration of survey.
Contributors
The following have contributed to this page
