Entry Name: "SMU-GUAN-MC1"

VAST Challenge 2017
Mini-Challenge 1

Team Members:

Guan Yifei, Singapore Management University, yfguan.2016@mitb.smu.edu.sg (Primary Contact)

Kam Tin Seong, Singapore Management University, tskam@smu.edu.sg (Professor)

Student Team: Yes

Tools Used:

Tableau

Link to the Tableau application created: https://public.tableau.com/profile/yifei2012#!/vizhome/Updated1/VASTChallenge2017MC1

SAS JMP (for data preparations)

Approximately how many hours were spent working on this submission in total?

80 hours

May we post your submission in the Visual Analytics Benchmark Repository after VAST Challenge 2017 is complete? YES

Video

https://youtu.be/i6BK2OxiaA0

Questions

1 – “Patterns of Life” analyses depend on recognizing repeating patterns of activities by individuals or groups. Describe up to six daily patterns of life by vehicles traveling through and within the park. Characterize the patterns by describing the kinds of vehicles participating, their spatial activities (where do they go?), their temporal activities (when does the pattern happen?), and provide a hypothesis of what the pattern represents (for example, if I drove to a coffee house every morning, but did not stay for long, you might hypothesize I’m getting coffee “to-go”). Please limit your answer to six images and 500 words.

Note: The anomalies identified in part 3 will be excluded in the patterns of life analysis as well for the same reason.

1. Setting the scene – classification of visitors & the “connecting path”

The visitors were divided into 6 segments based on the types of vehicles they drive and their activities patterns, summarized in Fig 1.A below. The patterns of life analysis will address each segment separately. The categorization allowed us to compare patterns between similar groups, eg daily campers vs those who came for extended period campings, to explore if they behave differently in the preserve.

Fig 1.B shows the most frequently visited paths (by gate categories) for all segments of visitors, in which we noticed the route “general gate-rangerstop-rangerstop-general gate” appeared very frequently. This is the only path that connects the east side and west side of the reserve for the visitors (excluding rangers), which explains the heavy traffic along this route. For easy identification, this path will be called as the “connecting path” in the subsequent analysis.

2. Daily trends for day campers

Refer to Fig 2-1, in general day campers started to enter the reserve at 5am and left the park by 18:00. In terms of camp selection, Fig 2-2 reveals that camp3 and 6 are their most popular choices, possibly because they are located nearer to the entrances. Accessibility could be one of the considerations for the day campers in choosing their campaign locations due to the limited time they have in the park.

3. Daily trend for extended campers

The day for extended campers started a bit later than day campers, from 6am onwards. The time of exit for them also spread over a longer time range (Fig 3-1). Unlike the day campers, they tend to settle down at the more secluded camp sites like camp 5 and camp 8, to enjoy a quiet night of sleep (Fig 3-2). For certain reasons, camp 1 was seldom considered as a camping location for the extended campers.

4. Daily trend for rangers

The day of rangers started from the ranger base and ends there with their first shift started at 6am and the last shift started at 17pm (Fig 4-1). They travelled by long but fixed paths, the reason could be that they are working on routine shifts (Fig 4-1). There are two typical paths adopted by the rangers, the maintenance path and the patrolling path. The maintenance path passed through various restricted gates, which was normally started at 6am,11am and 14pm. The patrolling path covered the entire reserve, started at regular hour intervals.

5. Daily trend for sightseeing coaches/visitors

The sightseeing coaches brought guests to the reserves throughout the day, without entering into the camp sites. Majority of the coaches transported visitors across the reserve, passing through the “connecting path” (path 1 in Fig 6). The rest of coaches revolved along entrance 1 and entrance 0, transporting the visitors along the camping path (path 2 in Fig 6) which connected camps 3,4,5

2 – Patterns of Life analyses may also depend on understanding what patterns appear over longer periods of time (in this case, over multiple days). 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. Characterize the patterns by describing the kinds of vehicles participating, their spatial activities (where do they go?), their temporal activities (when does the pattern happen?), and provide a hypothesis of what the pattern represents (for example, many vehicles showing up at the same location each Saturday at the same time may suggest some activity occurring there each Saturday). Please limit your answer to six images and 500 words.

Note: The outliers observed in part 3 were removed for the analysis of long period patterns to avoid misleading patterns.

1. Campers

Fig 7 provides the yearly visitor traffic calendar for the campers. The campers visited the preserve more often from May to Aug, possibly because this is the warm period of the year. The highest traffic of campers was observed in July 2015. There is a drastic drop in the campers, especially extended campers, from Q4 2015 onwards, which could be attributed to the colder weather.

2. Rangers

Fig 8 below reveals weekly activity pattern for rangers. In 8.1, the heatmap was configured to show the average stay duration for the rangers at various gates. We noticed the rangers would stay for extended durations at camping 8 (Mondays, 10am to 14pm) as well as gate 2 & rangerstop 1 (Mondays 6am – 11am, Wednesdays 13 – 16pm). The rangers could be doing inspection or maintenance works at this these locations. Looking at the reserve map, we can observe that ranger stop 1 and camping 8 are both located at the “dead ends” of the preserve, with no paths extending beyond them – it is likely that these two locations are surrounded with floras whereby periodic maintenance is required.

In Fig 8.2 we could see the rangers gathered at the rangerbase and gate 8 (which is in close proximity to the ranger base) on Thursdays 14pm. It could be an indication that the weekly ranger meetings were held on Thursdays 14pm at the rangerbase.

3. Service Trucks

Fig 9 below shows the weekly movement pattern for service trucks at various gates. We noticed that there were a higher number of service trucks moved pass the “connecting path” on Thursdays, at two prominent timings: 1am and 16pm. This might be the scheduled delivery/pick up hours for the service trucks.

4. Sightseeing coaches

Lastly, Fig 10 shows the weekly movement pattern for sightseeing coaches. The sightseeing coaches seemed to be bringing the visitors to the preserve on fixed days and hours, as the darker blocks on the heatmap tend to appear in regular intervals. For example, the coaches tend to visit the preserve at below timings:

- Thursdays: 11am

- Fridays: 3am

- Saturdays: 2am & 16pm

- Sundays: 3am, 11am and 16pm

3 – Unusual patterns may be patterns of activity that changes from an established pattern, or are just difficult to explain from what you know of a situation. Describe up to six unusual patterns (either single day or multiple days) and highlight why you find them unusual. Please limit your answer to six images and 500 words.

The Movement Anomaly

We first used the “Movement Anomaly” dashboard to discover the anomalies in the visitors’ movements. Each individual movement was represented by a Gantt bar. The Y axis contains all the days in the observation period and X axis shows the hours of the day. Filters allows the users to filter to see the activities at restricted gates, or only the activities by certain type of cars or visitors. Three movement anomalies were observed.

In 12.1, we filtered away the 2P cars and filtered in only the restricted gates and noticed two types of trespassing behaviours:

a. A group of 6 cars (type 1) entered from entrance 1 trespassing restricted area ranger stop 1 from 10am to 16pm, on 10 July 2015. The sensors at gate2 did not capture any of their activities, most likely they moved to rangerstop1 from the entrance1 directly through the jungles. As discussed earlier on, rangerstop 1 is one of the areas that are frequently maintained by the rangers and floras could be found there. This could be one area where the birds are nesting.

Car IDs	Movement
20152810102803-808
20152810102819-458
20152810102828-459
20152910102928-970
20152910102959-782
20153010103017-871

b. 4 vehicle entered from entrance 3 trespassing restricted areas (gate5,gate6,rangerstop6,gate3 & ranger stop 3) 23 times from 2am to 5 am, only observed on Tuesdays and Thursdays. The trespassing cars followed almost exactly the same paths. This looks like some planned acts which were only performed under the masks of the dark night. Type 4 vehicles are the heavy trucks; they could be transporting some illegal materials in or out of the preserve repeatedly.

Car ID	Movement
20150104020118-228
20150416040441-902
20150505020522-625
20150920030917-854
20151112031119-409
20151201031245-77
20151414041406-386
20151415031450-923
20151520021556-881
20151521021518-235
20152824032830-251
20152925022919-735
20153923043910-954
20154702044723-914
20154901044910-777
20154907044911-419
20155201025245-696
20162219032229-226
20162401032410-101
20162419042411-322
20163016033037-38
20164531024545-131
20165003035005-470

In Fig 12.2, we kept only the “extended campers” and included all the gates in the analysis. We noticed some suspicious movements of the extended campers at 0 hours which the rest of the extended campers would not be active at this hour. Interestingly, we noticed all the activity records belong to the car ID 20154519024544-322, which stayed in the preserve for 5 months.

Stay Duration Anomaly

In the “Stay Duration Anomaly” dashboard we introduced two scatter plots for anomaly discovery.

1) Extended campers:

· Abnormally high number of stops visited

a. Car-ID 20154519024544-322 visited total 281 stops in the park and stayed in the park from 19^th Jun to 5^th Oct. This is the same car we observed in 12.2

b. Car-ID 20154112014114-381 visited 98 stops and stayed from 14^th Jun to 26^th Jul 2015

· Abnormally high average duration in the preserve with low number of stops visited

c. 20150105060134-242, 20150420100416-232 visited 4 checkpoints but stayed for over one month in the reserve

2) Sightseeing visitors with abnormally long stay in the park, these are the same cars we observed in 12.1

Car ID
20152810102819-458
20152810102828-459
20152910102928-970
20152910102959-782
20153010103017-871

3) Service trucks with abnormally long stay in the preserve.

Car ID
20150204100226-134
20150322080300-861
20153427103455-30
20154501084537-684
20160623090611-424
20161008061012-639

4 –– What are the top 3 patterns you discovered that you suspect could be most impactful to bird life in the nature preserve? (Short text answer)

1) The trespassing service trucks we identified in 12.1. As explained, the movements by the trucks were the most suspicious among all the anomalies discovered. Further investigations are recommended to be conducted to find out what were they transporting in or out of the preserve – whatever it was, it would not be good for the preserve. Even the truck drivers were not hunting the birds directly, they might be transporting certain things that are harmful to the natural environments in the preserve, which in turn would affect the birds’ habitats.

2) The long stay duration car ID 20154519024544-322. This car recorded 281 visits to 10 gates, which shows a very suspicious loitering behavior. It could be moving around the preserve searching for something, could be the birds!

3) The movement of service trucks and sightseeing guests/coaches all night long. We observed that unlike the campers, the service trucks and sightseeing coaches move along the connecting paths all night long, going from one side of the preserve to another. The constant traffic at night might be a nuisance to the birds’ life, and caused the birds to migrate to other quieter places outside the preserve.

Besides the possible caused identified above, there is another area of concern – the fixed patrolling routine of the rangers. In our path analysis, we noticed that the rangers are working on fixed patrolling routes. It would be relatively easier for people with evil intentions to observe the patrolling routes and schedule, and plan their activities accordingly to avoid being caught. This might be a potential threat to the natural life in the preserve.