Tuesday, October 17, 2017

GIS 5935: Lab 7- TINS and DEMs

Good evening UWF,

This week's lab was fun and relatively easy playing with data in DEM and TIN format.  I have personally not worked with TINs a lot in my school work or career as of yet, so this was a good introduction for me into this realm.  The main focus of the lab was looking at TINs in ArcMap and ArcScene, and seeing different elevation characteristics associated with both.  I learned that TINs are a great way of showing detailed elevation characteristics in smaller study areas.  Visually, larger study areas are better having DEMs associated with them.  This lab had me deal with four of five different TIN datasets and I think that I enjoyed Part D, where I analyzed the difference in contours.  The contours were generated from both a DEM and TIN, and the differences can be viewed easily.  DEM contours are smoother, where TIN contours are jagged.  Usage of either depends on the user, but visually DEM contour lines look better.  A screenshot of that difference from Part D of the lab is shown below.  DEM contours are in black and TIN contours are in blue.


-Matt Griggs



Wednesday, October 11, 2017

GIS 5935: Lab 6- Location-Allocation Modeling

Good evening UWF,

This week's lab dealt with location-allocation modeling to help locate the best facilities to correct market areas to best serve and meet demands.  This type of analysis can be a little dangerous in my opinion because of all of the variation involved.  This lab dealt with making so many joins, exports, and summations that it was a little hard to keep track of everything.  I got stuck at one point and needed assistance via our class discussion post board.  Luckily someone was able to help me out and I was able move on.

The blog post map deliverable is displayed below.


-Matt Griggs

Tuesday, October 3, 2017

GIS 5935: Lab 5- Vehicle Routing Problem

Good evening UWF,

This week's lab dealt with determining the best route to be used to cover a certain area.  Our example included delivery trucks making certain stops throughout southern Florida.  I have never used Network Analyst in this capacity before, but I can definitely see the benefits from it.  The main benefit is all of the data linkages that are made between solving the Vehicle Routing Problem.  The Network Analyst essentially creates a series of join between stops, vehicles, and cost to provide excellent summary data on what is needed to complete a route based upon resources on hand.  This could be done for one vehicle and one route or a whole fleet of vehicles with multiple routes.

I would like to think of way of incorporating this into some of the GIS work that I do with my job.  My clients do a lot of inspection and maintenance work throughout Jacksonville.  Maybe one day a scheduling system can be implemented based upon sites visited on a specific day to see how many sites can be checked daily to increase efficiency.

Te require screenshot for this lab is displayed below.  It shows an adjustment made to the number of trucks sent out to complete the orders needing to be visited.  The original VRP produced 14 trucks with routes.  Multiple routes got missed though and there were numerous time violations that incurred through this system.  The adjustment of two more trucks increased efficiency by being able to meet all orders and routes with only one time violation.


-Matt Griggs

Wednesday, September 27, 2017

GIS 5935: Lab 4 - Building Networks

Good evening UWF,

This week's lab dealt with building transportation networks using Network Analyst in ArcGIS.  The complexity of this lab was staggering to me, as there are some many variables in play.  The user can have a lot of creativity in regards to setting up their transportation network within Network Analyst.  As an example for this lab, we played with allowing and not allowing restricted turns.  The results of finding the quickest route based upon these turns being allowed or not allowed has a strong effect on total time for the route to be complete.

No map deliverable is required for this blog post, but I do want to note that there was some confusion between professor and classmates on what geodatabase to use for this lab.  I used the San Diego geodatabase based upon a classmate's suggestion so I hope that is what we were supposed to use.  In any event, I was able to answer all of the questions asked in the lab based upon the data provided and created in the San Diego geodatabase.

-Matt Griggs

Wednesday, September 20, 2017

GIS 5935: Lab 3- Determining Quality of Road Networks Part 2

Good evening UWF,

This week's lab assignment continued off of the building blocks learned in lab 2 dealing with other road network quality analyses.  Lab 2 dealt more with analyzing road networks with points and intersections.  This lab dealt with analyzing completeness of road networks via their shape length.

I used some interesting tools in ArcGIS that I don't use very often during this assignment.  I used the Intersect tool to break road segments up based upon grid boundaries that they crossed.  This tool then keeps unique road segments of the same named road and splits them at the grid boundaries.  This splitting method was key to determining some questions asked in the lab.  Using the Summarize function from the Intersecting Grid layer that I created was able to show me unique totals of road length for each grid amongst my two compared road datasets.

The map deliverable for this assignment asked to show which grids had more complete road networks based upon the two compared datasets.  I used and "equals if" statement in Excel to find which grid had a higher measure value of road between the two datasets.  The high the measure of length value correlates to the better completeness.  Upon completing this analysis, I then joined this Excel table in ArcMap to the Grid layer and symbolized the grids accordingly.


-Matt Griggs

Wednesday, September 13, 2017

GIS 5935: Lab 2- Determining Quality of Road Networks

Good morning UWF,

This week's lab dealt with reporting accuracy in regards to road networks and their perceived intersections.  I liked that this lab required the user to analyze three different data sets to determine accuracy.  Two separate data sources were given to us as road layers.  The lab required us to create road networks from this data and create "junction points" at every intersection of the road networks.  A random sample was required to be collected from the two different junction points datasets, and from there the user then created their own junction point based upon aerial observation.  Then X and Y coordinates were created for the three junction point datasets.  That data was then uploaded into a provided accuracy spreadsheet that complied RMSE and NSSDA statistics.  The fact that this spreadsheet did this automatically was nice!  The results then allowed the user to create an accuracy statement to represent their data.  Accuracy statements were required for the City and StreetMap USA spreadsheets.  These two data sets were compared to the user created "Reference" junction points.  Their accuracy statements are displayed below.

City Positional Accuracy:  Tested 29.3 feet horizontal accuracy at 95% confidence level.

StreetMap USA Accuracy:  Tested 398.1 feet horizontal accuracy at 95% confidence level.

I really enjoyed this lab, and I feel that I can apply this to my work.  I will definitely keep this spreadsheet template handy for other geographical point analysis work.  The map deliverable required for this lab is just a screenshot of all of the Reference points created by me for this analysis.  I created 40 points on accident instead of the 20 listed in the lab.  The reference points are collected in a random stratified sampling pattern.


-Matt Griggs

Wednesday, September 6, 2017

GIS 5935: Lab 1- Calculating Metrics for Spatial Data Quality

Good evening UWF,

This week's lab dealt with analyzing the difference between accuracy and precision in data and how that is measured.  In general, accuracy refers to the closeness of a measured valued to a known standard value (e.g. the distance of a dart from the bulls eye).  Precision refers to the variation observed when measuring the same part repeatedly with the same device (e.g. a dart hitting near the same spot on the board multiple times).  Precision and accuracy are independent from each other, so data can exhibit one, both, or neither of these traits.

The map deliverable for this lab required a display of points and the accuracy and precision existing within these points.  An average position was determined from the point data, and acts as the center of the map.  All of the true data points are scattered across the map.  There are some instances of both precision and accuracy shown in the data.

Once the map was created, a subsequent step allowed us to see a reference point that a GPS unit was supposed to match to in measurement.  Vertical and horizontal measurements were given for this reference point, and in comparison to the average point position in the center of the map, there is significant error.  A difference of 3.78 and 5.98 meters exist between the reference point and average point between their horizontal and vertical values given respectfully.

-Matt Griggs