Implementation approach

eCommerce.ai Julia package provide a unified Analytics platform to support data analytical operations on all sort of Sales, Orders | Revenue Management, Procurement, Accounts Payable, Procurement including Vendor, Use, Freight, Misc Tax Accruals data to address complete Buy to Pay data wrangling operations.

This package will provide a complete Analytic Software package, which can be deployed as a bolt-on or independent application for all data extract, load, transformation, ad-hoc reporting & Analytics, visualizations and tooling to support Data Science, AI, ML predictive Analytics.

This package is intended for small, medium, large and very Big Organizations who require a Big Data Tools which can ELT i.e. Extract very large amount of structured and unstructured data, load data into a uniform platform such as RDBMS, Hadoop Data Lake or non-SQL environment.

Further, advance data transformation wrangling techniques can be applied to prepare data for operations reporting, data analytic, advance data visualizations, data science operations including AI, ML for predictions.

This package also show case reporting, visualizations to support real time, live reporting on all mobile, web devices.

eCommerce.ai takes a methodological business workflow approach (follow data) to solve this challenge.

Step 1:

At first, a detail analysis (much of the work) is done to understand, define end-to-end source to pay, order to cash, procure to sell business operations.
You will see, tons of examples included in this project, These examples resemble real life commercial good procurement to sales including payments, accruals, receiving and expenses etc.

Step 2:

Next, 3rd part IOT data like, local community events, holiday calendars, long weekends, weathers, climatic conditions, type of data is gathered.

Step 3:

Then all of this data is combined, cleaned and wrangled in a format which can be used in Analytics.

Step 4:

Then after, following Analytics is run and made available (in form of Jupyter | Pluto notebooks) for business operations, KPI Dashboards and Executive dashboards. These KPIs help business leadership take effective operational intelligence decisions.

Final deliverables

Ad-Hoc reports :    simple data queries
Analytics:          Self service reporting, analytics & visualization
Advance Analytics:  would | could | should
Predictive Analytics:   train, test and predict KPIs
Real time Analytics:    running analytics on real time data

Architecture

Proposed Architecture

Workflow

Business process workflow

ERD

Physical ERD DB Architecture

Explainable AI

Explainable AI