ISSS608 2016-17 T3 Assign CHIAM ZHAN PENG

VAST Challenge 2017: Mini-Challenge 1

Mistford is a mid-size city is located to the southwest of a large nature preserve. It has been discovered that the number of nesting pairs of the Rose-Crested Blue Pipit, a popular local bird due to its attractive plumage and pleasant songs, is decreasing! Provided several datasets, it is required to investigate the reason for the decrease of Rose-Crested Blue Pipit. In Mini-Challenge 1, traffic movement dataset is given to analyze patterns of life of vehicles through out the reserve, and detect unusual patterns that are potentially harmful to the birds. Questions raised are:

• Describe up to six daily patterns of life by vehicles traveling through and within the park.
• Describe up to six patterns of life that occur over multiple days (including across the entire data set) by vehicles traveling through and within the park.
• Describe up to six unusual patterns (either single day or multiple days) and highlight why you find them unusual.
• What are the top 3 patterns you discovered that you suspect could be most impactful to bird life in the nature preserve?

This webpage will guide you through my investigations and help to save Rose-Crested Blue Pipit!

Data Description

Entry gates are positioned at the Preserve entrances. Each vehicle receives an entry ticket at the gate and is assigned a vehicle class; the entry is recorded. The entry ticket contains an RF-tag that enables the Preserve sensors to pick up the passage of a vehicle through the Preserve. Each vehicle surrenders their entry ticket when exiting the Preserve and the exit is recorded. A .csv file containing data recorded from sensors around the Boonsong Lekagul Nature Preserve. A map containing the locations of roadways and sensors throughout the Preserve is also provided.

Data Preparation & Data Visualization

Data Preparation

1. Mapping the coordinates of each checkpoint and checkpoint type Each checkpoint is mapped using JMP Map creator on a 200x200 coordinates and labelled according to the 6 checkpoint types.

2. Set the Origin ("From") and Destination ("To") & Path and 1 visit or episode (Entrance and Exit) Each sensor record with timestamp indicate a checkpoint with correlate a origin, destination and path. Each path is labelled an sequential path number to identify the sequence. To identify a single visit or episode, two consecutive records at an entrance indicates an exit follow by a entrance which is a different visit or episode.

3. Calculate the duration from one point to another. Using the timestamp entry value and taking difference, duration is calculated using mdd hh:mm format and compute the time duration in hours format. To identify a single visit or episode, two consecutive records at an entrance indicates an exit follow by a entrance which is a different visit or episode.

4. Preparing the path intensity data format Each path count is calculated with an path ID and duplicated in another row for it to be visualize in Tableu as per link. http://onlinehelp.tableau.com/current/pro/desktop/en-us/help.htm#maps_howto_origin_destination.html

5. Preparing the chord diagram data format In order to create the chord diagram, each origin to destination path's percentage needs to be first calculated and labelled as color1 and color2 accordingly.

Data Visualization

Go through interactive visualization storybook to understand and discover patterns and trends. Simply select, filter and hover for more details!

• Using Tableau 10.3 published in Tableau public

Main workbook: https://public.tableau.com/profile/zac.chiam#!/vizhome/Book3_5732/VASTStoryMC1?publish=yes
Chord Diagram: https://public.tableau.com/profile/zac.chiam#!/vizhome/ParkChordv2/Dashboard2?publish=yes

Patterns of Life Analysis

Usual Daily Patterns

1. Movement hours and cars in the park: 8a.m to 5p.m peak hour

• People started to arrive in the park from 5 am onwards until 7 pm; with peak hour roughly between 8a.m to 5p.m.
• Even at quiet hours between 7 p.m. to 5 a.m., there are all types of vehicles that travels within all the gate types in the park
• Most common vehicle types are 2 axle car/motorcycle & 2 axle truck

2. Traffic behavior: General-gate is most accessed and silent hours of campsite and ranger-base

• General-gate are the most accessed checkpoints in the park
• Campsite: There is no movement at the campsite between 12a.m to 5a.m.
• Ranger-base: There is no movement of ranger-car and at the ranger-base between 3a.m. to 6.a.m.

3. 2 & 3 axles buses & 4x truck bypass through the park and do not camp

• 2 & 3 axles buses move throughout the park at all hours but not to camp sites. It passes through:
  • all entrances 0,1,2,3,4
  • some general gates 1,2,4,5,7
  • some ranger stops 0, 2, which is on the main route
• 4 axles truck move throughout the park at all hours but not to camp sites. It passes the same checkpoints as 2 & axles buses and in additional:
  • gate 3, 5 & 6
  • ranger stops 3 & 6

4. Time taken between public checkpoints and Rangerstops: < 24 mins and 1hr

• The average time taken to travel between checkpoints for public checkpoints (excluding ranger-stop andis under 0.4hrs or 24 mins.
• Rangers typically spend less than 1hr at all rangerstops except for ranger-stop 1 which has exceptions up till almost 6 hours.

Usual Multiple Days Patterns

1. Visitors during summer

• Most visitors during the summer (June – Sep) when it is warm in the park and peak month in July. As can see, the traffic is increasing in May 2016 as the weather is warmer and approaches summer.

2. More visitors during the weekend

• More people during weekends (2 axles car/motorcycle). Traffic increases on Friday till Sunday and drops when during when it approaches weekdays.

3 Weekend camping at the campsites

• The campsites are where the average time spent per location is highest meaning people spent most of their time at camping in the park. Average duration is about 2 days across most campsite except for campsite 1.

4. Multiple episodes visits: Transport vehicles & Weekend camping

• There are various multiple (more than 1) episode/visits to the park by mainly by 2, 3 & 4 axles truck probably as transport vehicles and 2 axles car/motorcycle for weekend camping at campsite 4, 6, 0.

5. North and central route heaviest traffic

• Path in the northern part of the park from gate 3 to general-gate2 and range-stop 0 & 2 than general-gate1 are the highest, followed by in the central of the park from general-gate1 to general-gate4, 7 and 2.

Unusual Patterns

1. 4 axles truck entering ranger-stops

• There are a total of 23 entries of 4 axles truck that goes to ranger stop 3 & 6 which is restricted area for rangers only, that travel on the same route path and time that avoids the ranger-vehicle.

2. 2 axles car/motorcycle entering ranger-stops

• There are a total 6 (20152416012433-808, 20152810102819-458, etc) of 2 axles car / motorcycle who also went to ranger-stop 1 which is restricted area for rangers on the same day 10 July 2015 at about 1035am. TThe route for both the 4 axles truck and 2 axles car are mapped out and this could be an suspected organized group activity.

3. Repeated (16 times) weekend camper moving in the late night

• There is a visitor, 20154519024544-322 on a 2 axles truck that repeatedly enters the park to campsite 4 and leave the park at past Sunday midnight or early Monday morning which is an odd hour to travel and possibility avoiding to be notified or seen.

4. Long-term resident camper

• There is a visitor, 0155705025759-63 that has been in the park since 5 June 2015 to 20 May 2016, of more than 11 months without leaving the park. Key question is on the source of food and possibly hunting on animals or birds in the park.

5. Vehicles travelling too fast and slow

• There are a couple of cars that are taking longer time than average between the routes that might be doing something other than driving. There are also some cars taking lesser time than average which means they are probably speeding above the speed limit of 25 mph.

Top 3 Possible Causes

• 1. 4 axles truck that enters ranger-stop 3 & 6 on multiple occasions that avoid the ranger-park vehicles operating hours. It is a heavy vehicle that can be used for transportation, or carrying equipment to perform quite a wide possibility such as deforestation that can potentially affect wildlife and cause for the decreasing bird population.
• 2. Long-term camping “resident”, 0155705025759-63 that has been in the park since June 2015, key question is on his source of food and activities at the various campsites.
• 3. Multiple episode/visits overnight camper, 20154519024544-322 that stayed at campsite 4 before leaving during the midnight. This is an unusual behavior that can be intentionally avoiding attention or being seen by others.

Comments & Discussions

Hi Zac, Really enjoyed your sharing! Below are some of my opinions, and you are very welcome to comment on my work also :) Aesthetic: Without proper title for the charts, it takes more time to digest the information that you try to present. Zac: Basically, I’m lazy but agreed; titles will be added. I am not sure color-coding the gate-names makes it more informative. Zac: It is to group them together but I do agree find it that to have 2 same graphs but different legends and colors different to follow but I just want to show car-type and gate-type information together. I think it will help investigation if selecting legends can highlight the information in the graphs. And don't you think some interactive techniques between different graphs will also help bridge the information together? Zac: Basically, I’m lazy but agreed; synchronized actions (filter, highlight) will help and will be added. I use box-plot to present duration as well, I think a horizontal lay-out makes it more friendly to read. Zac: Sure, will consider if it helps as there are too many paths. I like the idea that you plot the car-type as trellis, which makes the graph cleaner than color-coded them, and it's easy to compare! Love the chord diagram and it is very informative in showing the busy corridors! Would like to see a map next to it so that we know exactly where the corridor is. Zac: Good point and I think so as well to relate back to the actual map or route. Challenge is chord graph is ‘heavy’ and slow by itself (performance issue), filter action to a route on a map might make it slower but worth trying! Clarity: For 'number of records', is it based on number of episodes, or car-ids, or the number of rows? Zac: I assume you are referring to multiple visit/episode graph. It is based on number of rows or registered points on the sensor; main purpose is to identity which gate and car-type they are. I would like to know how you interpret the tree-map and how is it useful? Zac: It is same as above; It is to present overview of gate and car-type in a hierarchical manner. Of course, using maps/path or other graphs/charts like others is possible but this is to attempt a different visualization. Different information sharing same color scheme makes it a bit confusing (especially without a title to clarify), such as the first page of the dashboard. Zac: Agreed as the color scheme across dashboard has no intended relationship. Will add in the titles! Thank you, Joyce

Hi,Zac Great work! I enjoyed reading through your analysis findings. Yet, I have the following suggestions that hopefully are useful in further improving your work: Aesthetics: The chord chart really delivers good information. Yet the color is a bit confusing and the legends are not well attached to the graph. As a result, audiences need quite a bit time to understand it. Zac: Yes, there are just too many checkpoints in the park hence resulted in too colorful chord diagram. Perhaps removing the data from rangers will make it less clustered as but it is a first attempt at visualizing overview traffic and patterns. For the graph that shows the heaviest traffic in the preserve, all lines are a bit twined together, making it a bit difficult to read. I am wondering is there any better way to present the information. Zac: Yes, I have the same thoughts and challenges. I have already filtered out smaller paths to make it less clustered but I wish it can be clearer. The episode duration graph has about 40 box plots in a line, making it very difficult to read and compare.I am wondering if you can categorize these box plots using vehicle type to make it a bit easier to read and as well you probably can get more information. Zac: Good Point. I will breakup into car types to basically allow slice and dice. Clarity: I do not quite understand the tree map, not quite clear what messages you want to express. Would like to discuss more with you. Zac: Main purpose is to identity which gate and car-type they go to in a hierarchical manner. Of course, using maps/path or other graphs/charts like others is possible but this is to attempt a different visualization. There is lack of background information for some of the graphs. It is a bit difficult to interpret them for audience. For example, for chart (repeated 16 times), you did not say it is for car id "20154519024544-322". Audience who do not have a good understanding of the data would have no idea what you are trying to say. Zac: I have added titles to each of the chart and hopefully it’s better now. Good work! I enjoyed reading through your visual analysis. Hopefully my comments can add value. Thank you Yunna

comment3