Chapter 5. Anomaly Detection with R and Python

McKesson Corporation (McKesson), one of the nation’s largest distributors of pharmaceutical drugs, agreed to pay a record $150 million civil penalty for alleged violations of the Controlled Substances Act (CSA), the Justice Department announced today.

Department of Justice, January 17, 2017

Upon reading those headlines, Janine’s heart sank. She read the article with rapt attention; this affected her. She worked in the regulatory department at Big Bonanza Warehouse where she was responsible for maintaining corporate compliance. She was aware of Suspicious Order Monitoring Regulations (21 C.F.R. 1301.74(b)). Lately the Department of Justice was hitting companies left and right for noncompliance with this regulation, much more than they had done in the past. The regulation loosely states that companies that manufacture and distribute controlled substances “know their customers.” In the regulation’s exact words,

It is fundamental for sound operations that handlers take reasonable measures to identify their customers, understand the normal and expected transactions typically conducted by those customers, and, consequently, identify those transactions conducted by their customers that are suspicious in nature.

21 C.F.R. 1301.74(b)

But what exactly did this mean? She knew that her company had their sales orders in SAP. There were over 10 years of sales orders. But what did it mean to understand the normal and expected transactions

Get Practical Data Science with SAP now with the O’Reilly learning platform.

O’Reilly members experience live online training, plus books, videos, and digital content from nearly 200 publishers.