The multi-criteria evaluation and least-cost pathway analysis are applied to North-Western Ontario to determine the most suitable UAV launch site and access route through the application and reclassification of factors identified in literature, explained in objective 1, and outlined in Table 4.1. Further, determination of the specific weights assigned to criteria in this model are outlined in Table 4.2 with the final weights given in Table 4.3.
Table 4.1 Summary of criteria and implementation in MCE and LCP model.
| Component of Analysis: MCE |
Constraints |
Classification |
| |
Residential area |
exclusion zone including buffer of 1000m |
| |
Rivers |
exclusion zone |
| |
Slope |
>15 degrees exclusion zone |
| |
Roads |
exclusion zone |
| |
Evaluation Criteria |
|
| |
Land cover layer:
|
Classifications other than those described below considered suitable and givene value of 100
|
| |
Wetlands |
value 0 |
| |
Protected Areas |
value 0 |
| |
Lakes |
value 0 |
| |
Dense Forest |
intermediate value 50 |
| |
Distance from Thunder Bay |
multiple geodesic buffers standardized on scale of 0 to 100 with values descending with movement away from Thunder Bay |
| |
Distance from Remote Communities |
multiple geodesic buffers standardized on scale of 0 to 100 with values descending with movement away from remote communities |
| |
Distance from Roads |
multiple geodesic buffers standardized on scale of 0 to 100 with values descending with movement away from main highways 11, 17, 529, & 599 |
| Component of Analysis: LCP |
|
|
| |
Landcover Layer: |
Classifications in land cover layer other than those described below considered suitable and given cost of 1 |
| |
Forest |
cost of 50 (allowed but preferred to avoid) |
| |
Wetland |
cost of 7637 (determined by finding amount of cells on diagonal of map extent to create absolute barrier) |
| |
Residential Area |
cost of 7637 (cannot build) |
| |
Lakes |
cost of 7637 (cannot build) |
| |
Rivers |
cost of 7637 (cannot build) |
| |
Roads |
cost of 0 (already built) |
| |
Slope |
from 0 to 5 degrees, 1; from 5 to 10 degrees, 20; from 10-15 degrees, 40; and greater than 15 degrees, 7637 (Bell et al, 2002) |
Table 4.2 Pairwise Comparison Matrix
| |
Land Cover |
Slope |
Proximity to Preexisting Infrastructure |
Proximity to Thunder Bay |
Proximity to Remote Communities |
|---|
| Land Cover |
1 |
1 |
1/5 |
1/5 |
1/3 |
|---|
| Slope |
1 |
1 |
1/5 |
1/5 |
1/3 |
|---|
| Proximity to Preexisting Infrastructure |
5 |
5 |
1 |
3 |
3 |
|---|
| Proximity to Thunder Bay |
5 |
5 |
1/3 |
1 |
3 |
|---|
| Proximity to Remote Communities |
3 |
3 |
1/3 |
1/3 |
1 |
|---|
Table 4.3 Relative Weights
| |
Land Cover |
Slope |
Proximity to Preexisting Infrastructure |
Proximity to Thunder Bay |
Proximity to Remote Communities |
TOTAL |
| Land Cover |
0.07 |
0.07 |
0.09 |
0.04 |
0.04 |
0.07 |
| Slope |
0.07 |
0.07 |
0.09 |
0.04 |
0.04 |
0.07 |
| Proximity to Preexisting Infrastructure |
0.33 |
0.33 |
0.49 |
0.63 |
0.39 |
0.43 |
| Proximity to Thunder Bay |
0.33 |
0.33 |
0.16 |
0.21 |
0.39 |
0.28 |
| Proximity to Remote Communities |
0.2 |
0.2 |
0.16 |
0.07 |
0.13 |
0.15 |
| Sum |
1 |
1 |
1 |
1 |
1 |
1
|
Factors outlined in Table 4.1 are then applied to the MCE as outlined in Figure 4.1 to determine the most suitable site for a UAV launch port to service remote communities out of Thunder Bay. As well, Figure 4.2 shows how the variables from Table 4.1 are applied to the LCP to identify the least-cost path to connect the UAV launch port to Thunder Bay for transportation of goods to the UAV launch port for flight to remote communities.
Figure 4.1 Overview of Multi-Criteria Evaluation
Multiple suitable sites fit the area requirements. So, to determine the best site, the LCP analysis was actually applied to all three sites that were the outcome of the MCE to determine which one had the least cost to connect to Thunder Bay via pre-existing roads. This was then chosen as the most suitable site.
Figure 3. Overview of Least Cost Pathway
