Before the pandemic, ‘supply chain’ wasn’t exactly a household word. No one thought much about it until they couldn’t find their favorite yeast for the baking bread frenzy that ensued or toilet paper on grocery store shelves. During the pandemic, we have seen the impact a world hooked on social media has had on product sales, from skyrocketing demand that was unexpected, to rumors of shortages that placed inordinate stressors on production planning, output and logistics..  We also witnessed what happens when the supply chain just cannot keep up – or isn’t agile enough to use real-time data as circumstances unexpectedly change. This ‘black swan’ event of 2020 triggered a new level of consumer awareness, placing a spotlight on supply chain issues that are much bigger than just temporary supply shortages.

In a perfect world, data should empower organizations to have visibility across their supply chain, ensuring the right product and materials are at the right place at the right time. But this isn’t a perfect world, and Big Data from the supply chain, (often gathered over decades within businesses) simply has become too unwieldy for humans to decipher effectively and a lack of confidence in the data has made it essentially useless.

 To remedy the problem, companies traditionally have turned to data cleansing projects, the process of fixing or removing bad syntax or ineffective files to pare down to only the data needed for actionable and/or operational use. The data cleansing process usually includes a never-ending series of aggregating, organizing, analyzing and integrating dirty and duplicate material. This is difficult enough in each manufacturing facility or plant, much less across a disconnected network of facilities.

 It’s an understatement to say cleansing dirty data is expensive, takes too much time, too many resources or outside consultants, too much investment without a valuable return and has proven to be unsustainable…Yet, we all know it’s necessary.  Therein lies the challenge. You must do  something with improved data for it to make a difference in the business outcome whether it is process improvement, or delivering direct savings to the bottom line.

If there is a silver lining to the pandemic, it is that technology evolution and adoption has been forced to accelerate. One area that is having a major positive impact on the supply chain is harnessing the power of AI and ML to leapfrog the cumbersome process of traditional data cleansing.

The need for better business outcomes: without traditional data cleansing 

There has long been a need to use data more efficiently in the manufacturing supply chain, but this is an industry typically slow to adopt new technology. ‘Covid-19’ added a new layer of urgency to take a deeper look into how it can improve. While you can’t predict a pandemic, it has highlighted why organizations need more resiliency and agility in their supply chain. Having experienced what we have these last few months, it has become clear that if a supply chain is focused on becoming more agile, it will be more resilient reacting to real-time changes. 

Operating the supply chain in a Covid-19 world has shown that data is in a state of mess. Supply chain organizations weren’t able to spend several months cleansing data to ensure product production and delivery were meeting current needs. By not having control over the data, manufacturing wasn’t able to shift quickly to recognize what the immediate needs were in stores and warehouses weren’t able to adjust stocking and delivery processes.

These types of supply chain failures are still happening, and they are happening on three levels. First, there is the data problem. Data is dirty, redundant, and of low quality. There are disconnected siloed facilities unable to “speak” to each other. This impacts productivity in a negative way because it is inflexible.

Second, the traditional data cleanse process is manual, and can’t keep up with real-time changes. In fact, organizations overall have been slow to adopt supply chain technologies, resulting in increasing inventory costs and inventory that isn’t meeting current consumer needs. Multiple business units have different goals and incentives, so there is an overall disconnect on what data is needed for a streamlined supply chain process.

This leads to the third level—huge costs. Poor supply chain data management can cost millions a year in wasted working capital.

The pandemic not only showed the importance of a well-oiled supply chain process, but it has also served as the wake-up call that the data cleanse is no longer efficient or effective. The time has come to make the leap and take advantage of what AI and ML offer.

AI’s function in eliminating the data cleanse: speed and scale

Gartner has predicted that “By the end of 2024, 75% of enterprises will shift from piloting to operationalizing AI, driving five times increase in streaming data and analytics infrastructures.”

No organization has the luxury of waiting until 2024 to make the shift to AI for better data analytics. Covid-19 has made it clear that the need to implement a digital transformation was a year ago, but if that didn’t happen, then the transformation has to be moving forward now. No company can afford to fall behind because of delays in the supply chain.

This is where AI and ML come in. While there is still a tendency in the manufacturing and  supply chain industry to think of AI as new, it has been around long enough that there can be some familiarity and comfort level in its ability. In terms of its benefits to the supply chain, AI and machine learning are able to cut through the noise of Big Data. AI technology can now examine and sort the demand signals it receives and hone in on real-time information. The data analysis process is proactive and can cut through dirty data and react to immediate needs. Whereas a data cleanse can take upwards of nine to eighteen months, and can still require plugging the data into spreadsheets, AI cuts that cleansing period from 2 years to 2 months if applied correctly.  What was time consuming and messy data becomes quickly understood, available, intelligent, actionable data.

For companies that have been sitting on the fence, the time has come to make the decision to embrace AI and the proven benefits it offers in terms of visibility into data, eliminating duplications and optimizing inventory. Bottom line, AI allows for better decisions about data while creating actionable intelligence in real-time.. Given the spotlight on the supply chain and the insights we have seen in the last few months, can you afford not to evaluate and innovate?