Modern retail has seen something of a tech revolution in the past few years, with new innovations becoming available on a regular basis.
The pace of change is so great that today’s bright spark is next year’s standard requirement.
For example, the smart shelves launched by Mondelez International, in 2013 were marketed as state-of-art technology, but are now considered a simple modular solution (Dell, Panasonic – Smart Shelves). Such is the rapid evolution of technology.
Similarly, recent hot topics, such as cloud and big data, will soon no longer be considered cutting-edge technology. Indeed, the technology frontier and competitive edge will move towards the Internet of Things (IoT), and the shift has already started.
So, what is IoT? Gartner describes it as “the network of physical objects that contain embedded technology to communicate and sense or interact with their internal states or the external environment”.
A recent Gartner survey has marked IoT as the fastest growing trend in technology and estimates that by 2020, the number of Internet connected objects will increase by 30 times.
Many are familiar with Pervasive Computing/Ubiquitous Computing and will imagine IoT as just a new moniker to an old concept.
However, what has changed recently is the evolution of cloud technology.
Thanks to cloud-enabled solution offerings, actionable data can now be generated by connected objects.
With the decreasing cost of integrated chips, technology supported by a cloud platform and powered by big data analytics along with machine learning is the triad that is driving the growth of IoT.
While IoT represents a huge opportunity for almost every facet of human enterprise, this is especially true for the practitioners of supply chain/operations/analytics.
Leaders in both e-commerce and traditional retailers are exploring adoption of IoT to gain a competitive edge.
The aim is to gain better visibility along the demand and supply side of the value chain; to reduce variability by actively countering any ripples along the chain, and to get better returns by enabling optimised velocity of product flow.
There are several developments already underway in IoT that are revolutionising the traditional retail supply chain in multiple ways right now.
All supply chains originate with the customer and the most exciting area in this field is the innovative integration between the end user and IoT.
The primary focus here is to gather customer data, customise assortments/offerings to individual needs provide an attractive value proposition and simplify the buying process.
All of these devices are constantly collecting data about customers’ preferences, influencers etc. and enable an unprecedented opportunity for razor sharp product positioning.
For instance, recommendations can be provided about a nutrition bar based on a person’s surfing history, palate preferences and social media influences and can also adapt these recommendations based on whether a customer has joined a health centre or purchased a fitness tracker, etc.
One of Microsoft Bing’s latest developments has seen the analysis of large samples of search queries to predict internet users who may be susceptible to pancreatic cancer, even before they have been diagnosed with the disease.
Imagine a retailer being able to deploy this technology on its website.
So, what about retailer’s decision making? There are some equally interesting cases for the use of predictive analytics for businesses intelligence.
At the retailers’ end come smart shelves and the retail store. Current category management processes and its tools work by taking sales data as the main demand signal to build forecasts/simulations and associated analytics.
In the future, this is set to be revolutionised further. As an increasing number of IoT devices are deployed with blockchain technology, we are set to see vast advancements in independently thinking retail infrastructure.
By effectively recording a ledger of data exchanges between devices, the web as well as consumers, blockchain and IoT come together to provide retailers a significant competitive edge.
For example, imagine a smart shelf that knows the purchase habits of its customers, can not only monitor and
report on its own stock – but also bid distributors for stock replenishment as well as pay for the delivery of those new items automatically.
With smart shelves in place, retailers can accurately track OOS by triggering an event in case of an empty shelf or by actively tracking shelf-fill levels.
Digitally linked price tags can also allow seamless changes in prices from online to in-store and enable an omni-channel product offering.
To continue with our nutrition bar example, time spent in front of a particular product category (say, low fat yoghurt) can be an early indicator for a change in recommendation/ promotion.
Further evolving into an integrated retail IoT, the shelf can be configured to generate orders for itself.
Indeed, the whole environment can be configured to access a planogram library, the store inventory data and tie up with material handling bots (automated warehouses) to execute automated shelf-fills at the backend.
While the store is a more recent development in the application of IoT, both transportation and warehousing were its first frontiers; the implementation of RFID heralded the first generation of machine-driven dataset.
Integrated tracking systems and sensors have been in use for a long time in warehousing and transportation networks.
RFID-tagged pallets have also allowed deeper visibility in inventory status and location.
With convergence of demand signals and greater visibility into inventory status and location; we have a scenario where Amazon has obtained patent rights on “Anticipatory shipping“.
Increasing integration promises that the IoT will eventually result in effective use of material handling bots (mentioned above) as well as delivery by drones.
These growing innovations are challenging the effectiveness of current network optimisation systems by providing machine learning as an effective alternative.
Machine learning relies upon data collected from integrated and interfaced objects/sensors and combines it with optimisation using a distributed file system.
This makes it ideal to continuously optimise an ever-changing environment and learn from it.
The overall evolution of IoT environment has been a meandering journey with moments of serendipitous delight.
Even with all the advantages that it promises to deliver to businesses, IoT remains a very challenging venture with big risks and unsolved problem areas.
For any organisation that has decided to embark upon IoT, there are open questions on all key fronts.
Several interfaces work very well in particular focus areas, but there is a lack of standardised platforms.
Industry experts have floated Platform as a Service (PaaS) as the next area to enable support to a growing IoT environment.
Despite these challenges, technology seems a hurdle that can be crossed, the merging of processes with human interactions will have a steeper curve.
Also, it is important to remember that although IoT is ground-breaking, simply implementing this technology will not suffice.
Retailers and solution providers need to work together to ensure that this data is deconstructed, analysed and implemented in a way that enables greater customer satisfaction and meets business objectives.
Sourced by Vivek Wikhe, senior solution advisor, JDA