Entry Name:
“UBA-Rusconi-MC2”
VAST Challenge 2016
Mini-Challenge 2
Team Members:
Ivo Rusconi, University of Buenos Aires, ivorusconi@yahoo.com.ar PRIMARY
Dawoon Choi, University of Buenos Aires, dawoonchoi330@gmail.com
Pablo Martinez,
University of Buenos Aires, pablowmartinez@hotmail.com
Student Team: YES
Tools Used:
Power BI
Microsoft SQL Server Management
Studio
Excel
Approximately how many hours were spent working
on this submission in total?
120 hours.
May we post your submission in the Visual
Analytics Benchmark Repository after VAST Challenge 2016 is complete? YES
Video
Questions
MC2.1 – What are the
typical patterns visible in the prox card data? What does a typical day look like for GAStech
employees?
GAStech employees typically work from Monday to Friday. Time shifts depend
mainly on the department they work. For instance, Facilities have three different
shifts covering every hour of the day, each day of the week. Engineering and
Information Technology have two shifts, whereas the rest of the departments
work mostly from 7/8 to 17.
Below we can see some
examples:
Executives are the ones
with a more flexible schedule, not having a strict hour to enter and exit the
building. Opposed to this department we can see people from Security who are
the most exact and punctual. These differences are surely related to the kind
of job they have and the tasks they accomplish.
Each person from each
department spends more time in a specific place of the building and we can see
some patterns or clusters associated to them.
Below we
can see an example from 2nd floor:
Also
interesting to visualize through a graph:
·
The central node represents
the department and the other nodes represent floor-zone.
·
The thickness represents
the time consumed in each of the places.
MC2.2 – Describe up
to ten of the most interesting patterns you observe in the building data.
Describe what is notable about the pattern and explain what you can about the
significance of the pattern.
As commented before, it is
interesting to see how Gastech employees spend their
time at work. How different departments have different habits and schedules. We
find this remarkable especially if we were interested in studying and
understanding people’s work and relations. We found it difficult sometimes to
difference the events we came across between a regular pattern,
an anomaly or an unusual event since the time window observed was quite narrow,
and sometimes the lack of a pattern could be considered a pattern itself. In
consequence, the reader will sometimes find some answers to be cross different
questions. The study of patterns is interesting, but it is however much more
interesting when it is broken.
We found five people from
the Administration department curiously enter the building every day at the
exact time each. For instance, cforluniau gets to
work every day at exactly 7.55 or eklinger enters at
8.30 each and every day. Gflorez, jfrost
and mbramar have the same conduct. One possible
answer is that they have perfectly arrange their schedule and commuting so they
get each and every day at the same time. However, in a real world it is
difficult to see this and not think that there’s some error or manipulation
with the data.
Below is an example about cforluniau where we can see she enters the building (1st
floor-zone 1) every day at 7.55 AM:
We
discovered that zone 9 from floor 3 is the area where most energy is spent.
This is probably due to the fact that the server is located in that area.
Supply inlet mass flow rate is also highly superior than
in other areas, probably because of the same reason.
MC2.3 – Describe up to ten notable anomalies or unusual events you see in the
data. Describe when and where the event or anomaly occurs and describe why it
is notable. If you have more than ten anomalies to report, prioritize those
anomalies that are most likely to represent a danger or serious issue for
building operation.
The first important thing
we found is that there is an employee tracked who is not in the Employee List.
Although it is probably due to an error in the making of the list, we consider
it worth remaking. His ID is morlunv and we thing
he/she would probably belong to the Facilities because of the similarity to
another ID. The ID vmorlun is present in the Employee
List but not in the records from the building. Since this was found through the
analysis and the preprocessing of the data we couldn’t found an appropriate
visualization to show it although it is worth mentioning.
One important anomaly
involves gflorez and jsanjorge.
Even if they both regularly enter around 8am, there is a check at 0am. Gflorez the 7/6 and jsanjorge the
2/6 checked in the building at that hour. There is no information about them
close before or after that hour. This may probably be due to an error in the
registry or some other strange event.
Below we see an example
about Gflorez:
Although people usually
work from Monday to Friday we can see lcarrara, llagos, mbramar and ostrum entering the building during a weekend in different
moments. Mbramar and ostrum
both get to work on a Saturday (11/6) and nearly at the same hour. It could be
explained as extra work, although it is something remarkable.
Below we see an example
with Mbramar:
We saw seven people who
have their last check inside the building in areas not related to the elevator
(zones 1 or 4). While the “check in” of the following day does not present any
anomaly, which is an anomaly itself.
Below is an example about edavies who’s last check (31/05/2016 11:59:00 PM) was
registered on 2nd floor zone 1:
In
addition, there is one employee (fresumir) who
“enters” the building one day in zone 2 of floor 1. And here again, there is no
anomaly present the day before.
Below is
the mentioned case:
There are some offices (or
zones) where no prox-card detections are registered
with the robot. Those included:
·
Zones 1,2,3,4,5,12c,12b
and 16 from 2nd floor :
·
Zones 1,2,12,11c and 11b from
3rd floor :
It can be explained as
normal if those areas were not in use but it is probably not the case of all of
them.
We can see Mat Bramar from Administration, the 31/05, checking in zone 5
from floor 3 which is indicated as a “future expantion”
area. As so, we think people should not be allowed to enter there.
It is highly remarkable that there are only five pairs of (X,Y) coordinates detected by the robot in floor 1:
MC2.4 –– Describe up to five observed relationships between the proximity
card data and building data elements. If you find a causal relationship (for
example, a building event or condition leading to personnel behavior changes or
personnel activity leading to building operations changes), describe your
discovered cause and effect, the evidence you found to support it, and your
level of confidence in your assessment of the relationship.
The zones where Hazium is registered (floors 2 and 3) don’t have prox-card registries from the robot. This is probable
because of the fact that Hazium is probably a
dangerous gas and personnel don’t enter in direct contact with the element.
2nd floor:
3rd floor:
There are no prox-cards
detected by the robot in zone 2, floor 2, where there is also a high
concentration of CO2 on the day 10/06 so it could be related to a specific event.
Another relation, also considered an
anomaly, is that on the 02/06 there is an incoherent temperature in zone 16,
floor 2.
Finally, we found some patterns in relation with the reheat coil power,
summarized as below:
·
3rd floor:
o 31/05 there’s no registry.
o 01/06 is only registered in zones 5 (without detections by the robot) and 6.
o 02/06 is only registered in zone 1 (without detections by the robot).
o For the rest of the days (except on Saturdays and Sundays where the zone
with the highest average is the 8th) zone 1 presents the highest average.
It is remarkable that there are never detections of prox-cards
in zone 1.
·
2nd floor:
o There are several days where no measuring is registred.
·
1st floor:
o There are several days where no measuring is registered.
Curiously, except 06/06, the
days match where no measuring is registered on 1st and 2nd
floors.