Real-Time Supply Chains and the Quest for More Orange Soda

Real-time supply chain orange soda

The notion of real-time supply chains that dynamically serve the needs at points of demand with waste-free, precise planning and execution is certainly the goal of every B2B manufacturer. The legacy approach to managing these supply chains involve any number of centralized systems and the aggregation of demand across what might be dissimilar points of consumption opportunity. The complexity of how to get to that goal of a real-time from the current state can be daunting and unclear. 

A simpler example can provide insights into approaches that can be applied to even the most complex supply chains. We don’t need to look any further than the vending machine in our lunchroom to find an evolving real-time supply chain.  Historically a vending machine was located near someplace where it seemed like people might like to purchase and drink a beverage. The vending machine took real cash (remember cash?) and required regular visits from an attendant with a cart full of replacements. This system served its clients for decades but was fraught with efficiency issues.

The soda machine may be conveniently located for the consumer but not so conveniently located for the service personnel. This means that the attendant may have trekked down to the basement of a dorm with a set of replenishment soda that does not match the most recent consumption.  He may have brought with him too many cola replacements and not enough orange soda. Not all orange soda customers will settle for a second choice resulting in lost revenue.  Even worse the attendant may show up with inventory only to find that the machine had a mechanical problem and that it has been out of service for some period of time. This failure results in wasted service time, missed demand and unhappy customers that found other ways to satisfy their beverage needs. 

The individual characteristics of each vending machine were often lost in the aggregation of data across the entire fleet.  The average consumption of orange sodas per week may not reflect that there were just a few machine locations that disproportionately drove that consumption.  This results in too much orange soda inventory in some machines and not enough at the heavy consumption points. Servicing demand with an “average” supply builds inefficiency into the supply chain.

In today’s world of IoT, the “connected” vending machine has come on-line.  The modern vending machine can automatically and immediately report the current situation at each point of demand.  The overall status of the machine, the current inventory of each product, the temperature of the machine, the last time the machine was serviced and a number of other criteria are readily available.  The modern vending machine takes no cash which eliminates a mechanical point of failure at each machine, reduces the risk of handling cash, improves cash flow and provides more accurate revenue data.

There are some obvious benefits in terms of servicing this modern supply chain. Many of the inefficiencies of the service of the supply chain disappear.  The attendant now arrives with a replenishment set of inventory that exactly matches the needs at the individual vending machine eliminating the disappointment of orange soda lovers. The knowledge also allows a single replenishment attendant to service more machines as he can skip regular visits to machines that are not in an immediate need. The attendant can also prioritize machines that are reporting issues and restore them to working condition to avoid lost opportunities. 

The access to richer and deeper data for the evolved supply chain needs to be evaluated and acted on at both a centralized level and at the point of consumption level.  Centrally applying this layer of data collection and automation would optimize the current replenishment process and likely drive savings in-service personnel effectiveness and incremental sales by avoiding lost opportunities. Overall inventory in the supply chain can be reduced as clarity around real-time consumption and replacement needs become more apparent across the entire supply chain.  Economies of scale in terms of purchasing products will also occur as volume discounts can be requested for the faster moving product. 

Insights into individual vending machine locations can be evaluated at the central location.  Why does a small set of locations require more orange soda than cola? What does that say about customer demographics?  Are there additional products that can be offered to match those demographics?  Not every machine should have the same inventory and the customization of the supply chain can result in significant growth opportunities. 

Eventually, the vending machines will become intelligent service agents of their local demand.  There are already vending machines with sensors for outside temperature and even some with biometric sensors that become aware of the number of people that pass by and their heartbeats.  These machines may be capable of making independent decisions that further enhance the real-time support of the supply chain.  If the temperature is particularly warm in a location then the demand for a cold beverage is accentuated.  The modern vending machine can independently adjust the price of the product upwards to take advantage of consumer willingness to pay for the product. The vending machine could also reduce the price of an offering to avoiding losing a sale.  If you want an orange soda but the machine is temporarily out of that inventory could the machine offer you an alternative at a reduced price to assure that you buy now?  All of this location-specific knowledge can allow the individual machines to use machine learning to develop behavior that best serves their specific location. 

This supply chain example emphasizes the need for developing access to real-time demand data. It is critical that this data be evaluated to deliver both centralized macro insights and clarity around individual elements of the supply chain that allow for personalization of supply to individual points of consumption.  Can the concepts in this simple example be applied to your B2B supply chain?  Or have I just made you want an orange soda?  See you at the vending machine!