Small Data vs Big Data for Supply Chains

By Bernard Milian

Instead of Big Data, Why Don’t We Take Care of our Small Data First?

Big data and advanced analytics are front page news in the supply chain and have been on the rise for years in the vocabulary of consulting firms and software vendors. 

Accumulating masses of information, passing it through the mill of statistical algorithms or even artificial intelligence, detecting weak signals to identify trends, all this is very attractive. 

However, when it comes to managing inventory sizing and replenishment, determining the right priorities, and thus feeding into our teams’ day-to-day decisions, we need above all some basic data of our own, on which we can rely. 

The reality of many businesses today, large and small, is that basic planning data is often – hum – fragile. 

In the course of the many DDMRP implementations that we have accompanied, it is striking to note that decades after the advent of ERPs, the mastery of key supply chain data is not there, and the first step of a DDMRP project is often to re-examine the data: 

  • Define realistic procurement or manufacturing lead times 
  • Understand what lead times are expected by customers 
  • Have clearly defined lot sizes, mini / maxi / increments 
  • Define what is deemed to be available-to-stock or not 
  • Make sure no promise date is in the past (won’t happen) 
  • Maintain reasonably accurate stock records 
  • Keep BOMs up-to-date 

This is often not for lack of having taken care of the subject when implementing the ERP, and often having called in consultants, sometimes several times, to help squaring the system.  

My interpretation is that the problem is due to a lack of focus. There is so much information in an ERP system, so many things to keep up to date, that after a while you forget the essentials, and you get used to working with data that you know is wrong. 

DDMRP is more tolerant than MRP of data inaccuracies, particularly unreliable forecasts. However, it is a matter of managing an operating model by computer, and therefore the digital representation of the physical model must be reasonably good. 

The good news is that DDMRP makes it much easier and quicker to get the baseline data right, because anomalies are immediately obvious! 

The few important data are clearly identified, without getting lost in the maze of the ERP. 

The design phase requires the definition of deadlines and stock positioning.  

Visual management does the rest: If a 2019 sales order balance remains in the system, the dark red execution alert cannot be missed, if an incorrect lot size results in a huge green zone and disproportionate target stock, you cannot miss it.  

From experience, in a few weeks the system is back on track. 

Best of all, the continuous improvement management inherent in DDMRP ensures that this key supply chain data remains correct over time. 

So if you are offered yet another initiative to repair and rework your MRP, then switch to DDMRP now to fix this once and for all!  

Get in Touch

Share This Story, Choose Your Platform!

Share on facebook
Facebook
Share on twitter
Twitter
Share on linkedin
LinkedIn

Recent Posts

You may also enjoy

Stay informed with our latest updates

Name*