KPI:

In terms of the project deliverables, we need a “dashboard app” to display the data we have collected and worked on for this project. In terms of that, the KPI’s following is related to this:

  1. Number of Leads
  2. Cost per Lead
  3. Cost per Acquisition
  4. Social Media Traffic
  5. Social Media Conversion Rate
  6. App Store Download Number
  7. Sales Revenue

DATA SOURCES:

  1. NYC Open Data Restaurant Dataset:
    1. https://data.cityofnewyork.us/Health/DOHMH-New-York-City-Restaurant-Inspection-Results/43nn-pn8j
    2. https://data.cityofnewyork.us/Health/DOHMH-New-York-City-Restaurant-Inspection-Results/rs6k-p7g6
  2. NYC Open Data Rodent Population Dataset:
    1. https://data.cityofnewyork.us/Health/Rodent-Inspection/p937-wjvj

DIMENSIONAL MODEL (Draft):

  1. Requirements – MD
    1. Document overall Enterprise business processes (According to Gartner)
      1. Strategy to execution: Delivering organizational goals in the form of tangible projects
      2. Business process analysis: Optimizing specific process to gain efficiency and effectiveness
      3. Enterprise architecture (EA): Holistically analyzing, planning and modeling business structures
      4. DigitalOps: Technologically planning, modeling, coordinating, governing and monitoring the processes and resources associated
    2. Identify ONE core business function to model
      1. Integrate the functions and new data coming from the data sources
    3. Identify source OLTP systems that support/feed this core business function
      1. OLTP systems provide source data to us.
      2. UPDATE and DELETE data once a week when we’re updating the data warehouse.
    4. Gather business user requirements in terms of KPIs
      1. Need to create a social media integration to the dashboard to share the analytics
      2. Ability to handle analytical errors
  2. Grain – Pete
    1. Identify the basic unit of fact data (grain) for the data warehouse
      1. Restaurant_name,cuisine_description, Inspection_Date, inspection_type, Job_ticket or work_order_id, violation code, location
    2. Identify the summary grains for the data warehouse
      1. Rodent population by violation code and location
    3. Identify preliminary dimensions to support the grain
      1. Inspection date, job_ticket_or_work_order_IF
  3. Dimensions – Luan
    1. Identify all data warehouse dimensions:
      1. Location, Date, Grade, Restaurant, Inspection, Restaurant would each be a dimension set up for a star schema around one or two fact tables
    2. Identify slow-changing and fast-changing dimensional attributes
      1. Type 4 SCD, because although inspections happen at a low frequency, and restaurants rarely change locations, they will happen frequently enough so that creating a dimension for historical data would benefit the analytics processes.
    3. Level of granularity
      1. These fact tables will have a low granularity because we have the specific longitude and latitude as well as daily dates.
  4. Facts – Steven
    1. Identify facts
      1. Restaurant Name
      2. Cuisine Type
      3. Inspection Date
      4. Inspection Type
      5. Description
      6. Job Tickets
      7. Food Orders
      8. Violation Code
      9. Location
      10. Critical Flag
      11. Grades
    2. Identify fact types (additive, etc.)
      1. Semi-Additive
        1. Job Tickets
        2. Inspections Types
        3. Violation Code
        4. Inspection Dates
    3. Identify accumulative facts
    4. Design composite keys for fact tables
    5. Estimation of fact table size and growth