IS428 2017-18 T1 Assign Victoria Koh Wei Ting

Overview

Mistford is a mid-size city is located to the southwest of a large nature preserve. The city has a small industrial area with four light-manufacturing endeavors. The Rose-Crested Blue Pipit is a local bird popular for its attractive plumage and pleasant songs. Recently, there has been signs that the number of nesting pairs of the Rose-Crested Blue Pipit is decreasing. The decrease is sufficiently significant that the Pangera Ornithology Conservation Society is sponsoring researchers to undertake additional studies to identify the possible reasons.

The four factories in the industrial area are subjected to higher-than-usual environmental assessment, due to their proximity to both the city and the preserve. Gaseous effluent data from several sampling stations has been collected over several months, along with meteorological data (wind speed and direction) that could help us understand what impact these factories may be having on the Rose-Crested Blue Pipit.

Suspicion

Despite the four factories being quite compliant with recent years’ environmental regulations, there are doubts that the actual data has been closely reviewed. The noxious gases pouring out of the smokestacks from the four manufacturing factories south of the nature preserve are suspected to be the main contribution of the downfall of the Rose-crested Blue Pipit.

The Task

General Task

To visualize the given datasets and determine which (if any) of the factories may be contributing to the problems of the Rose-crested Blue Pipit.

Specific Tasks

Task 1

• Characterize the sensors’ performance and operation. Are they all working properly at all times?
  
• Detect any unexpected behaviors of the sensors through analyzing the readings they capture.
  

Task 2

• Determine which chemicals are being detected by the sensor group.
  
• Observe the patterns of chemical releases as being reported in the data.
  

Task 3

• Determine which factories are responsible for which chemical releases. Carefully describe how the deduction is made using all the data available.
  
• Describe any observed patterns of operation revealed in the data for the factories identified.
  

Data Analysis & Transformation

The following is the .xlsx files I have received:

1. Meteorological Data.xlsx
2. Sensor Data.xlsx
3. Sensor Location.xlsx

Before we visualize the data using Tableau, data cleaning is required to ensure all data works properly.

Data	Problem	Solution
Meteorological Data.xlsx	Missing rows and rows with missing values Redundant Elevation column	Remove rows 445 and 460 Remove column E
Sensor Data.xlsx	Sensor Data.xlsx has hourly readings while Meteorological Data.xlsx has 3-hourly records.	Grouped Date Time of sensors to match Meteorological data
Sensor Location.xlsx	No coordinates of the factory location.	Create new .xlsx file called Factory & Sensor Location.xlsx containing coordinates of the sensors and factories

Data Import Structure & Process

Process	Description
Importing and joining tables	Left joined Sensor Data and Meteorological Data
Wind direction angles and factory coordinates	Creating custom shapes of the wind arrow angles and factories using Photoshop and adding them to Tableau Repository

Interactive Visualization

The interactive visualization can be accessed here: https://public.tableau.com/profile/vick#!/vizhome/assignment1_62/MistfordVisualDetectiveStory

Interactive Techniques and Visual Guides

For the best experience, adjust your screen resolution to 2560 x 1600 and enable full screen on the browser. To help users navigate through the different filters and actions so that their analysis can be performed smoothly, the following interactivity elements and visual guides are added:

Interactive Technique	Rationale	Brief Implementation Steps
Month Selector	Allow the viewer to view the overall observation across all three months and be able to zoom in to the specific months.	Drag Date Time to Filters card and select Month / Year. Right click on the MY(Date Time) filter pill and select Show Filter. Go to the filter card and select the small arrow at the top right, go down the list and select Single Value (list).
Chemical Selector	Allow the viewer to view the overall chemical concentration observation and be able to zoom in to the specific chemical types.	Drag Chemical to Filters card, select All' and click OK. Right click on the Chemical filter pill and select Show Filter. Go to the filter card and select the small arrow at the top right, go down the list and select Single Value (list).
Sensor Range Slider	Allow the viewer to view the overall sensor observation and be able to zoom in to the specific sensors.	Drag Sensor to Filters card, select All' and click OK. Right click on the Sensor filter pill and select Show Filter. Go to the filter card and select the small arrow at the top right, go down the list and select Single Value (slider).
3-Hour Time Range Slider	Allow the viewer to explore the wind direction observations through 3 hour time periods.	Drag 3Hr Grouped to Filters card, select All' and click OK. Right click on the ''3Hr Grouped filter pill and select Show Filter. Go to the filter card and select the small arrow at the top right, go down the list and select Single Value (slider).

Visual Guide	Rationale	Brief Implementation Steps
Number of records color reference	Allow the viewer to be able to identify the areas with none or duplicate records.	Drag Number of Records to Marks card under the Color label. Configure the color settings to Custom Diverging and change the colors to white on the left end and red on the right end. Check Stepped Color and key in 3 Steps. Check Start and key in 0, and End and key in 2.
Chemical legend	Allow the viewer to be able to identify the 4 chemical types.	Drag Chemical to Marks card under the Color label.
Average Chemical Reading legend	Allow the viewer to be able to identify the chemical concentration readings.	Drag Reading to Marks card under the Color label.

The Story

The story mode is broken down into 5 “chapters” - Introduction, Sensor Pattern & Records, Chemical Release Overview, Chemical Release Focus, and Wind Direction Overview. The user should follow through the modes from right to left going along with the number tasks for Sensor performance and operations, Chemical concentration amount and records, and Factories’ chemical release relationship with environmental factors (i.e. Wind) .

Story Introduction

Sensor Patterns & Records Dashboard

Chemical Release Overview

Chemical Release Focus

Wind Direction Overview

Analysis

Task 1: Sensors’ performance and operation

The 9 sensors measuring the 4 chemical concentrations (AGOC-3A, Appluimonia, Chlorodinine, Methylosmolene) each function largely across 3 months (April, August, and December), with readings logged 24 hours per day at hourly intervals. There are however, missing data for certain chemicals on certain days.

Observation 1: Missing data

1. Consistent missing data at 00:00 at the first of each month across all months and all chemicals

Based on the record frequency graph, I observed that the missing data occurred on dates:

• 2 April 2016
• 6 April 2016
• 2 August 2016
• 4 August 2016
• 7 August 2016
• 2 December 2016
• 7 December 2016

2. Numerous duplicate data for AGOC-3A for sensors 3, 4, 5, 6, 9

3. Numerous missing data for Methylosmolene for sensors 3, 4, 5, 6, 9

4. Duplicate data pattern for AGOC-3A is the same as missing data pattern for Methylosmolene for sensors 3, 4, 5, 6, 9

Observation 2: Sensor patterns

1. Generally stable patterns in sensors 1, 2, 7 and 8
Sensors 1, 2, 7 and 8 have some variations in their noise but have generally stable and horizontal trendline.

2. Pattern anomaly in sensors 4, 5 and 9
Sensor 4’s noise shifts linearly across the months, giving a diagonal trend line. Sensors 5 and 9 get noisier over the months, with sensor 5 having a gradual noise increase and sensor 9 getting significantly noisier in the second month.

3. Pattern anomaly in sensors 3 and 6
Sensors 3 and 6 have more variations in noise compared to the other sensors.

Task 2: Chemical concentration amounts and records

Observation 1: Readings in certain chemicals

1. Significant readings for AGOC-3A and Methylosmolene on average per day across all 3 months

2. Unusual Methylosmolene readings from 22:00 to 04:00

Based on this graph, we can see that there are significant readings in AGOC-3A and Methylosmolene on an hourly average across the 3 months. However, Methylosmolene appears to have an unusually high average between the hours of 22:00 to 04:00 across the 3 months. Note: Methylosmolene was mentioned that factories have been required to discontinue using it for environmentally safer chemicals.

References

• Tableau Help - Create Advanced Highlight Actions (http://onlinehelp.tableau.com/current/pro/desktop/en-us/actions_highlight_advanced.html)
• Tableau Public – How to Make Control Charts in Tableau (https://public.tableau.com/en-us/s/blog/2013/11/how-make-control-charts-tableau)