Mars has been on a journey to transform their digital supply chain. In 2019, Will Beery was appointed the vice president and tasked with leading the company’s global digital supply chain transformation. Mars is one of the largest fast moving consumer goods companies in the world. This private, family-owned business is headquartered in the US. The company revealed last year that they have revenues of $45 billion. The company now has 12 brands that generate more than a billion dollars in revenue annually.

Mars’s Digital Transformation

Mr. Beery started by asking, “what does a digital supply chain mean for us?” For Mars, digitalization has focused on improved customer service, cost reductions, and making their supply chain associates more productive. But how to do this was  very ambiguous.  Many initiatives could be undertaken, but not every initiative would be the right fit for Mars. Mr. Beery began by visiting Procter & Gamble, UPS Logistics, the devices division at Microsoft, and other companies with advanced supply chains. Mr. Beery was also visiting startups that had interesting technology for which the use case was not immediately apparent.

Mr. Beery concluded that the best transformation projects would be, one, improved supply chain planning. The second goal was to work with a fourth party logistics (4PL) partner that had invested in cutting-edge transportation management and visibility solutions. A 4PL is a company that plans and executes transportation on behalf of their customers. The final goal was to develop end-to-end visibility based on leveraging data analytics.

The goal of an SCP implementation was to effectively balance demand and supply. Here, Mars Wrigley  leveraged using Kinaxis for both demand and supply planning. Kinaxis is a proponent of using network planners for both supply and demand planning rather than having the planners operate in silos. Mars took that recommendation.

Mars Wrigley was particularly interested in working with a SCP supplier that could integrate artificial intelligence (AI) to support demand sensing. Demand sensing involves the use of the external data sources – particularly the latest sales and market data – to improve short-term forecasting and then be able to use that improved understanding of consumer behavior to improve their supply planning.

In terms of finding a 4PL solution provider, Mars wanted a company that could engage in truck loading and route optimization, and schedule shipments into and out of their warehouses. The 4PLs are not engaged in planning for intercompany shipments between distribution centers. Mars Wrigley chose Transplace, now UBER Freight, for North America and a partnership with Maersk/E2Open for ocean freight. The company chose 4Flow to do transportation planning and execution in Europe.

What a Supply Chain Digital Transformation Means 

Unsurprisingly, a company as large as Mars has a highly complex supply chain involving global sourcing, manufacturing, and distribution. “You know,” Mr. Beery said, “we’re quite proud of the supply chain. It’s been highly resilient over hundreds of years.” But the supply chain challenges associated with the pandemic drove home the need for increased agility. While the pandemic challenges are winding down, there are increased geopolitical issues to deal with and a potential recession. The need for agility remains. “We need resiliency, but we also need agility. We need to have the best of both. That is where developing the digital muscle comes into play.”

At the same time Mr. Beery was visiting multinationals with advanced supply chain practices, he was also visiting startups that had interesting technologies. One of those companies was Aera Technology. Initially, Mr. Beery found the technology intriguing, but he was not sure how to leverage their solution. Several potential use cases were examined before Mr. Beery landed on using Aera to improve supply chain agility.

The Journey Begins

Mars Wrigley began working with Aera in 2020 in the US. Mars Wrigley is the confectionery division of Mars. This division has well-known brands like SNICKERS, Skittles, and M&Ms. The initial focus for the Aera solution was the chocolate business.

Aera Technology was seen as a convergence point for what Mars Wrigley was doing in data analytics and AI with what the company was doing in supply chain planning. “We knew we were going to get value out of next generation enterprise planning,” Mr. Beery said. “But it wasn’t enough to do what we needed in that 0-to-12-week horizon” – the short-term planning horizon. “I was being asked to build a lot of data sets and analytics to try to understand (which facilities would) have out of stocks, or where we were going to have to cut orders.” For the largest retailers, short shipping results in fines and penalties.

“I’ll be frank, it was hard to get humans to understand what we were trying to do. The change management was difficult.” The goal was to take data directly out of disparate source data systems including different versions of SAP and different types of supply chain planning, and then quickly assemble a common data model. Once the data was assembled, Mars Wrigley coded algorithms on to Aera analytics platform to detect problems and find solutions. The solution was then fed right back into the source systems that the data was pulled from. This led to automated execution surrounding the predicted supply chain problems. Aera refers to these algorithms as Aera Skills. The Skills implemented included stock rebalancing root cause  analytics skills.

The stock rebalancing skill is designed to enable Mars to optimize DC to DC shipments. The algorithms generate recommendations on optimal ways to rebalance inventory based on cost and service level impacts. Cost considerations include making sure truck utilization is high and ensuring that items that are closing in on their expiration date get shipped out. Because decisions are written back into the stock transfer order portion of the SAP enterprise resource planning system (ERP), the solutions are easier to execute.

Aera also captures whether the action was taken. The business gained the ability to monitor, track, and audit whether the decisions were executed. Every day, as changes in demand and supply occur, Aera recommends the stock that needs to be moved from an origin facility to a new destination facility.

The root cause analysis skill was developed to identify reasons for customer service level cuts and revenue losses. This exercise was previously a cumbersome manual exercise done on a weekly basis for the top 45 stock keeping units (SKUs). The analysis was not all that accurate. “The reason this was a really complex thing for our associates and supply chain to understand was because they were relying on reason codes within the ERP.” These codes were put into SAP after an order cut had occurred. There were so many different things that could be the root cause of a short shipment that mistakes were inevitable.

Now, the Aera Skill ingests daily data on order cuts and then assigns order cut reasons for all SKUs. Aera has visibility to what SKUs are in their network of distribution centers (DCs) and whether the right inventory is in the right location to fulfill the orders that need to be shipped out of that DC. This visibility is key to both the root cause analysis and the inventory rebalancing.

Mars Achieves Stellar Results

There is a cost to rebalancing inventory. But Mr. Beery pointed out that once the inventory is rebalanced, and the reasons the inventory ended up in the wrong distribution centers is understood, the model – or digital twin – at the heart of the SCP system becomes better. Those rebalancing costs are greatly reduced going forward.

“It is only in the last 18 months,” Mr. Beery explained, “that I’ve seen that people believe the system, they understand it, and that they’re excited by this capability.” To the planners, it initially looked like Aera was duplicating what was being done by the Kinaxis supply chain planning system. Once they understood that shorter term planning was better when augmented by Aera’s solution, planner buy in began. “Now we’re operationalizing it. For a long-time we were just trying to prove it to the supply chain associates.”

Mars Wrigley is achieving success in North America with this solution. Planners are far more efficient. Truck utilization went from the mid-80s up to over 95%. 95% is very high performance on this metric; It implies significant transportation cost savings.  This also equates to less road miles and lower carbon emissions which helps Mars progress on their sustainability objectives.  Meanwhile, customer service levels improved by 2%. That 2% gain may not sound like much, but the increase in customer service increased revenue and is incredibly important for Mars’ in meeting their customers’ requirements.   Perhaps an indication that Mars Wrigley’s transformation succeeded is that Mr. Beery has since moved on to be the global chief information officer for Mars Wrigley.