Demand sensing: Artificial Intelligence and the 21st-century supply chain

demand sensing future insights network.jpg

Demand sensing combines big data and artificial intelligence to push supply chain planning into new levels of capability. In Part 1 of a special report, Johnson & Johnson’s Neil Ackerman and Nitza Pierce explain how companies can get started

In this ever-changing and highly connected world, how we adapt to rapid change (and thrive) is the question at the core of almost every commercial strategy. For supply chain planners, this demands a new level of performance to meet 21st century consumer expectations.

The only way to navigate and succeed in this competitive terrain is to provide the right product or service, at the right time, in the easiest, fastest and most personalized way possible for your customer.  Of course, without overstocking. Successful execution of ‘right place, right time’ is challenging enough, never mind the added challenge of providing the right product in a market inundated with new business marks, when accuracy is the most substantial influence on success.

The most elusive and important change in supply chain will be the ability to predict the future accurately – predictive demand. Traditional planning and forecasting techniques are slow, unresponsive, relatively inaccurate and based on legacy technology whose days are numbered.


New rules of engagement

To support the ‘now culture’, Lean may no longer lead in supply chain strategy. We need a strategy to support demand unpredictability, because all operational success is affected by accurate forecasting. The effects it has, from your bottom line to employee scheduling, mean nothing escapes the results of your forecast, so it makes sense to focus on your ability to sense demand.  

When forecasting demand, many organizations rely on semi-automated predictive software. These programs have robust algorithms that interpret patterns from historical sales data to predict future demand. But with the amount of change or loyalty attention deficit, and the lack of e-commerce data available to be truly effective, we need more effective ways to predict the future when the past may no longer have the same relevance. Another layer to the flaws of these outmoded methods is these algorithms typically only have one or two sources of information.  

Execution Always Counts

If service and customer experience are critical aspects of your offering, e-commerce is making life easier for consumers. With growth rates of 30%, brands are often out of stock because managing the growth is so tough – if you don’t win the online buy-box because you’re out of stock, you can’t win the sale. And when consumers come back online, they go to the last item they bought, not your item. Loyalty is to whoever had their product in stock.

As the field of business forecasting develops, technological improvements allow companies to experiment with more advanced concepts. These include Machine Learning, Artificial Intelligence, IoT, Voice Enablement, and now, Demand Sensing. This approach has been around for more than a decade but has been steadily gaining interest as the significance of accuracy and insight increase. Because this is the competitive battleground, even leaders in forecasting accuracy must look at new opportunities to maintain their dominant position as #1 in market share, efficiency and cost-savings.

While the term is not new, its application in the era of AI and big data is undoubtedly pushing new boundaries

Value creation through demand sensing

The rapidly developing science of demand sensing uses machine learning, new levels of processing power and vast swathes of data to create more accurate demand forecasting models. While the term is not new, its application in the era of AI and big data is undoubtedly pushing new boundaries. It is a phenomenal opportunity for the supply chain to further its case as a revenue generation activity with a major seat at the table rather than a cost center in the business.

Demand sensing gets you on the path to predictive demand by taking historical data, combined with future predictions of external factors. This really is predicting the future - using years of data derives an algorithm with an output that tells you what demand will be in the future, given a set of circumstances.

For supply chain planners, this allows their role to shift from batches, manual activities and Excel spreadsheets, to strategic and data-driven scenario planning for the future. For the consumer, this goes even further, because it allows us to be in stock at the right time at the right location and to avoid the damaging implications of out of stock, which leads to consumers buying another brand, or leaving your portfolio of brands. Anyone in the supply chain, sales or marketing world, would agree that if you know an accurate precision of demand, you will be at a competitive advantage forever.

Getting started on demand sensing

Starting down the path of demand sensing requires some fundamental steps in terms of data, people, technology and skills. From that base, it is possible to build and iterate as you refine capability and effectiveness. Key first steps are:

  • Bring together three to five years of accurate data for the past

  • Identify the numerous external factors that could impact your product demand in the future (these vary but might include climate, economy, population etc.)

  • Bring in the databases of forecasts for the future (or create them yourself)

  • Take your historic forecasts and assess how accurate they turned out to be, which gives you the deviations from the forecast - the bias.

  • Apply that bias to your most recent predictions.

  • Weight an algorithm between the past and the future to create your first sensing model.

  • Measure how you did.

demand sensing ackerman.jpg

When incorporating the external factors influencing the product and market, it’s critical that companies get the right data. Start with a big list (which can be as many as 100) then whittle them down to the most important, otherwise it will be a disaster and you’ll be in a data swamp. That’s why it’s critical to keep running strategic scenarios – run it, check the result and if you’re way off, you have probably picked the wrong factors. Keep experimenting with different factors until you find the right match, which is why you need such heavy data platforms - you are churning a tremendous amount of data as you run the scenarios.

Building the right skills and teams

Demand sensing requires significant upskilling. It’s a new science for supply chain that’s being taught now in some of the best universities like MIT, but finding the skills is a challenge. It requires data scientists, people with deep supply chain and demand planning backgrounds, and strong business sense to understand the markets where these products play, so you can ensure you capture all the external factors that influence your brand.

It needs cross-functional teams that draw skills from across the business, because to make this work you have to understand that supply chain is not a cost centre – it is part of the revenue generating machine. It’s another example in the modern era of significant technology skills and supply chain skills coming together as organizations have to transform their IT to support supply chain.

To the future generations data science skills will soon be like typing.  If you don’t have a certain level of comfort you will be at a real disadvantage.

Demand sensing: systems and software

Most organizations do not have the computing power to put this together -  many are held back by legacy ERP systems that are slow and antiquated, and the worst possible platform to execute demand sensing.

Demand takes curated and uncurated data, combines it, cleans it and generates an output. ERPs do not do that. There was a time when ERPs were all the rage, but if I was starting a company now, I would strictly use EPR for transactions and financial reporting. I would use an enterprise data grid for all my data platforms, so I can call on data and analytics much faster. That approach will increase as we head towards 2020 – and will allow companies to future-proof more effectively to capitalize on the next developments in data analytics.

To move successfully into demand sensing, technology options include outsourcing to new platforms, building their own enterprise data lakes, or using it in the cloud.


Neil Ackerman is the Senior Director, Global Supply Chain Advanced Planning, for Johnson & Johnson across all segments including Pharmaceuticals, Biomedical Devices and Consumer Products. He is responsible for accelerating supply chain innovation and enablement of advanced planning processes and technologies worldwide. His team is critical in bringing value-based prototyping to life.

Nitza Pierce is the Senior Manager, Advanced Planning, for Johnson & Johnson as part of the team accelerating supply chain innovation worldwide across the Pharmaceuticals, Biomedical Devices and Consumer Products segments. Prior to her current role, Nitza has held multiple positions within end-to-end global supply chain.