AI & ML in Urban Forestry New England ISA

Presentation at New England ISA on the use of Artificial Intelligence and Machine Learning in Arboriculture and Urban Forestry.

1. Branching Out: Integrating
Al & ML Into Arboriculture &
Urban Forestry
Josh Behounek
Davey Resource Group

2. Part 1:
A brief introduction to AI & ML

3. Artificial Intelligence
Computer
Vision
Robotics Natural
Language
Processing
Machine Learning
Supervised
Learning
Unsupervised
Learning
Reinforcement
Learning
Neural
Networks
What is Artificial Intelligence?
1950: Ability to sense, reason, engage and
learn
1980: Ability to learn from data, without
being programmed (Decision Trees, NN)
2010: Neural Networks with multiple layers, used
to solve very complex tasks
Deep
Learnin
g
1990: Ability to learn from very vast amount of
data and solve simple tasks

4. High-level Principle of Machine Learning
Data Model+
Training
Data
Machine Learning AlgorithmFeed Into Trains Inference Model
Test
Data
Real
Data
Prediction / Result

5. How do machines learn from data?
Data with labels Data without labels States + Actions
Mapping Classes Actions
Input
Output
Supervised
Learning
Reinforcement
Learning
Unsupervised
Learning
It’s an
apple!
These
are
apples
Model
Model

6. Two main groups of machine learning models
1 2
DOG
“analyse”
“create”

8. Software & Hardware Constraints

9. TURN INTO THE SKID

12. Make Your Life Easier

13. Part 2:
AI & ML Tools in
Arboriculture &
Urban Forestry

18. Street Tree Insights with
ChatGPT
Kent, OH

20. What are the top 5 insights from this
data?

21. Create presentation quality charts from
this information in separate jpeg format.

22. What management
recommendations
would you make
from this data?

23. Create a heatmap of
the tree condition.
use a blue and red
scale for this. blue
being poor condition,
red being good
condition.

24. Create an
optimized route
to inspect all
high risk trees
by walking

25. Create a
map of this
data using
public roads.

26. Write a bid specification
request for planting 30
shade trees along the
public right of way in
Bristol, VT.
[Your Name]
[Your Organization]
[Address]
[City, State, ZIP Code]
[Email Address]
[Phone Number]
[Date]
[Recipient's Name]
[Recipient's Organization]
[Address]
[City, State, ZIP Code]
Subject: Bid Specification Request for Planting Shade Trees in Bristol, VT
Dear [Recipient's Name],
I am writing to request a bid specification for planting 30 shade trees along the
public right of way in Bristol, VT. Our organization is committed to enhancing
the aesthetics and environmental quality of our community, and we believe that
planting shade trees will contribute significantly to achieving this goal.
Project Details:
Location: Bristol, VT
Number of Trees: 30
Scope of Work: Planting, Maintenance, and Monitoring
Deadline for Completion: [Specify a preferred timeline]

27. Write a bid specification
request for planting 30
shade trees along the
public right of way in
Bristol, VT.

1. Tree Selection: Provide a list of suitable shade tree species that are native to
the Bristol, VT area, taking into consideration factors such as growth rate, size
at maturity, tolerance to local environmental conditions, and aesthetic appeal.
2. Tree Sizing: Specify the size of the trees to be planted, ensuring they are
appropriate for the chosen species and in compliance with any relevant local
regulations or guidelines.
3. Planting and Installation: Outline the planting procedures and techniques
that will be employed to ensure successful tree establishment. This should
include details on site preparation, proper planting techniques, staking (if
necessary), and soil amendment requirements.
4. Maintenance and Care: Describe the recommended maintenance and care
practices, including watering schedules, mulching, pruning, and fertilization.
Specify the frequency and duration of maintenance visits required for the initial
establishment period and subsequent years.
5. Warranty: Provide details on any warranties or guarantees offered for the
trees, including replacement policies in case of failure within a specified time
frame.
6. Monitoring and Reporting: Explain how the project progress and tree health
will be monitored, and include a reporting mechanism to update the
organization on the status of the trees after planting.
7. Experience and References: Include information about your organization's

28. Write a bid specification
request for planting 30
shade trees along the
public right of way in
Bristol, VT.
7. Experience and References: Include information about your organization's
experience in similar projects and provide references from previous clients for
verification purposes.
8. Budget Proposal: Please present a detailed breakdown of costs associated with the entire project, including tree purchase, planting, maintenance,
monitoring, and any additional charges or fees.
Please submit your bid specification by [deadline for bid submissions] to the
following address: [your organization's address]. Should you have any
questions or require further clarification, please do not hesitate to contact me
via email or phone [provide contact details].
We look forward to receiving your bid specification and assessing the potential
for collaboration on this significant project. Thank you for your attention to this
matter.
Sincerely,
[Your Name]
[Your Title]
[Your Organization]

30. Slide courtesy of Jennifer Wills

31. Precautions
Do Not:
•Seek legal advice
•Upload private information
•Rely on medical information
•Financial planning
•Trust blindly
•Ignore bias

42. Analysis & Recommendations

43. Analysis & Recommendations

44. Sonic Tomography

45. Analysis &
Recommendations

46. Smart Tree Inventories

47. Step 1: We capture cm-accurate point cloud and
automatically identify each tree.
Tree
Tree
Tree

48. Step 2: Create a 4D Digital Tree Twin of each tree
4D DIGITAL TWIN
Multispectral Satellite Images
Panoramic Images
Point Cloud

49. Step 3: We analyze each tree and extract information
Clearance Issues
Live Crown Ration
% Dieback
Ecological Benefits
Digital Tree Twin
Change Over Time
Cohort Analysis
Size (DBH, Height, etc)
Species
Leaf Area Index
Leaning Angle

50. Step 4: Define outliers
Absolute Outliers (cohorts)
●Dead trees
●Too much lean
●Leaf Area compared to Size
●Canopy Width vs Tree Height
Relative Outliers (filtering)
●Trees > x”
●Trees in certain neighborhoods
●Certain species of trees
●Trees >40% dieback

51. Step 4: Define outliers
Absolute Outliers (cohorts)
●Dead trees
●Too much lean
●Leaf Area compared to Size
●Canopy Width vs Tree Height
Relative Outliers (filtering)
●Trees > x”
●Trees in certain neighborhoods
●Certain species of trees
●Trees >40% dieback

52. In Field
25 - 100%
Remotely
0 - 20%
Step 5: Davey Arborists assess outliers

53. Make the Right Decision,
on the Right Tree,
at the Right Time

54. Base Information – Species, Size, etc.

55. Health & Vitality

61. Road Clearance

62. Traffic Sign Clearance

63. Utility Line Clearance

64. Tree Protection Zone & Fall Zone

65. Human Thermal Comfort

66. Machine Learning
Advantages
●Objective
●Repeatable
●Efficient
●Precise

67. Removing Implicit Bias

68. The Future Is Now!
Josh Behounek
[email protected]
573-673- 7530