Badi Leonel, Instituto Tecnológico Buenos Aires, lbadi@itba.edu.ar
Meola Franco Román, Instituto Tecnológico Buenos Aires, fmeola@itba.edu.ar
Puente Julieta, Instituto Tecnológico Buenos Aires, jpuente@itba.edu.ar
Strubolini Diego Martín, Instituto Tecnológico Buenos Aires, dstrubol@itba.edu.ar PRIMARY
Student Team: YES
Tableau
Excel
Approximately how many hours were spent working on this submission in total?
36 hours
May we post your submission in the Visual Analytics Benchmark Repository after VAST Challenge 2016 is complete?
YES
Questions
MC2.1 – What are the typical patterns visible in the prox card data? What does a typical day look like for GAStech employees?
A GAStech employee starts his workday at 7 and ends it at 17, as it can be seen in Figure 1.
Figure 1: Number of Employee detections by hour for all departments from the proximity cards sensor
If we go further in detail, we can see that these hours vary according to the employee department. Specially if the work in the "Facilities" department, we can see that they have different schedules, since there has to be employees at every hour in that area. Also if we see the "Information Technology" and "Engineering" departments there are few peoples that stay as guards from 17 till midnight as it can be seen in the Figure 2.
Figure 2: Number of employees grouped by department by hour according to the card sensor
An average employee has his workplace in the second floor. This led to analyse Figure 3 and see that in the zone 1, the only zones where a large number of employees are register are the transit areas such as the zone 1 and 4 or the zone 2 that has a “Deli” (see Figure 4).
Figure 3: Number of employees by day, zone and florr according to the card sensor
Figure 4: Second Floor zones diagram
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.
1 - The Hazium concentration highest peak for the three floors coincided on June 11 after 18 hs. As it can be seen in Figure 5, in the first floor was at 18 hs, in the second at 18:05 hs and in the third at 20:50 hs. It is from this translation of the hazium that we can infer that this gas has a lower density than the one of the air since it goes up through the floors. Despite the fact that this gas goes up, it can be seen that the highest hazium concentration peak happened in the second floor. This could be generated by insufficient ventilation in that floor or a design error.
Figure 5: Hazium concentration average for the three floors
2 - In Figure 6, it can be seen that on Sunday the Water Heat Gas Rate that heat the water is the lowest of all week. This could be caused by the low number of employees in the building.In a lower measure it can be seen that Saturday also consumes less Gas. On the other hand, from Monday to Friday, the Gas Rate is constant.
Figure 6: Water Heater Gas Rate by Hour grouped by weekday
3 - In Figure 7 that the water temperature inside the tank shows little variations, which is a common behaviour in these tanks since these are normal values. These variations are not caused by the Water Heater Setpoint since it has been always constant.
Figure 7: Water Heater Setpoint and Water Heater Tank Temperature by hour.
4 - We can show a direct dependency between the average Return Outlet CO2 peaks for the entire building with the following attributes:
For this, the total average by hour was calculated for the entire building. The June 7 and 8 peaks from the first chart are replicated in the six following charts (All shown in Figure 8). And all the analysed variables are related with air changes.
Figure 8: Air related average in the entire building for six attributes
5 - The following attributes average was constant across all the days (See Figure 9):
Figure 9: Average values for the entire building for seven attributes that were constant
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.
1 - We can see that certain employees of the administration department entered the building on Sundays (See Figure 10). Also this behaviour is similar to the executives one but they work extra hours only on saturday. Maybe they are having an affair ? Are they plotting something? Or maybe they are just working extra hours ?
Figure 10: Number of Employees by department and hour on Saturday and Sunday.
2 - The registered water temperature entering the boiler has its highest value during night hours and the lowest ones in the middle of the day. This makes no sense because solar energy is at its highest point at noon (See Figure 11). This can be caused by a bad configuration in the inner clock of the sensor. This anomaly is repeated every day (See Figure 12).
Figure 11: Supply Side Inlet Temperature by hour for a regular day
Figure 12: Supply Side Inlet Temperature by hour for every day.
3 - We can see that the daily energy consumption of the "Deli" has a strange peak between 11 and 12 hs on May 31 (See Figure 13). Was this due to the connection of an extra coffee machine ? Or maybe the employees are using the electrical grid of the building for something else?
Figure 13: Equipment Power by hour for the Deli zone
4 - It is remarkable that the energy consumption in the third floor is much higher that in the other floors (See figure 14) . Also this energy consumption remains high in the weekends. We can explain this because a zone in the third floor is under construction and the construction equipment consumes a lot of energy. Or maybe there is an area that is being used for something else.
Figure 14: Equipment Power Average by day for each floor
5 - Another interesting aspect is shown in Figure 15. There are people who appear in the list of employees but were never recorded by card sensors. Moreover, it is important to mention that almost all work in the security department. Are they up to something? They could also be on vacation or the company has not updated the list of employees. However, this qualifies as an anomaly.
Figure 15: Never sensed by proximity sensors employees List grouped by department
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.
1 - In Figure 16 you can see that the Return Outlet CO2 concentration in the 3rd floor in the server room had two peaks on June 7 and 8, during which it reached a value greater than other days. This could have affected the employees working in that area, as you can see in Figure 17, no employee was sensed on the other day in the server room (June 9). This could be because CO2 has negatively affected these employees, preventing their attendance at work. This is unusual because it is a working day where you can see that in the previous week, people were sensed in that room.
Figure 16: Return Outlet CO2 Concentration in the Server Room zone by day.
Figure 17: Number of employees by day in the Server Room zone.
2 - In Figure 18, the Bathroom Fan Power for the first floor is shown. The first thing we can see is that during the weekend, the fan reaches its maximum power, so we can expect that the fan is mostly used when there are few people in the office (as explained in MC2.1) perhaps with the purpose of cleaning it. You can also see times on weekdays where the fan power reaches its maximum values in its power close to noon times, this leads to Figure 19 where you can see the number of employees and location on the floor 1 in a time slot from 10 to 14 hours. You can see that in the near bathrooms sensors, the number of employees is considerably lower than elsewhere. This could mean that the bathrooms are less crowded at these times and therefore the fan is more used.
Figure 18: Fan Power for the first floor by day
Figure 19: Number of employees in the first floor between 10 and 14 hs
