Idea

We spent a significant amount of time researching about the most effective ways to solve real world business problems in the supply chain industry. The technologies we use are all Open Source and can be learned within a relatively short period of time.

We believe that there is no need to purchase expensive software for completing tasks that could be done by the company’s internal IT department.

Our goal is to give other people a better understanding of artificial intelligence by sharing practical business cases we have worked on in the past.

The key takeaways from all our talks are:

Artificial Intelligence is not rocket science!

It is easy to setup solid applications by using software that is entirely Open Source.

Always look at problems from a business perspective, then work backwards to find the technologies.

There is no need to be perfectionistic, if the AI model yields a better result than what can be done manually it is a good starting point. The fine-tuning can be done later.

Datacrag Rocket

Introductory Talks

Following are introductory talks we held to help people in Hong Kong getting familiar with the topic Artificial Intelligence and Machine Learning.

How to get started with AI?

Artificial intelligence is a broad topic and it is easy to get overwhelmed by the amount of literature.

The purpose of this talk is to give a basis understanding of what artificial intelligence is and how it can be categorized.

How to get started with AI?
How does and AI get smarter?

What is the different between supervised, unsupervised and reinforcement learning?

This talk is an extension of the first one and is made for business people who want to enhance their technical knowledge.

How does and AI get smarter?
What to do with my data?

How to narrow down the data before running your models?

Here we want to help future Data Scientists understanding their data. Detecting outliers and cleansing the data are essential steps before running a model.

What to do with my data?

Business Cases

Following are some of the real world business cases we have shared in the past.

Clustering as a base for assortment planning

Example of a textiles manufacturer selling its products across 300+ retail-stores and an e-commerce platform.

Clustering as a base for assortment planning
Image classification for vendor allocation

Example of a sourcing company acting as middle-man between brands companies and manufacturing vendors.

Image classification for vendor allocation
Classification for controlling the material ordering flow

Example of a footwear manufacturer with a supply base of 80 manufacturing suppliers and 400+ raw material suppliers.

Classification for controlling the material ordering flow
Logistic regression as an optimization technique for fresh food replenishment

Example of a global service provider in the travel industry operating over 50 countries and 300+ airline customers.

Logistic regression as an optimization technique for fresh food replenishment