The Future of Onboard Inventory: How AI-Driven SCM Strategies Could Revolutionize Inventory on the Move
By land, by air, by sea – transit organizations know the competition is fierce for retaining customer loyalty, requiring inventory management strategies that balance consumer needs with sustainability initiatives and business goals.
These companies understand that one bad experience could jeopardize future sales.
Unlike other goods and services, a passenger on an airplane, cruise ship, or train cannot simply go somewhere else if you don’t have the soda or the meal that they want. And depending on their travel itinerary, that could become a very long trip for the passenger.
At the same time, companies are looking to be more environmentally responsible, reducing waste by not carrying more inventory than necessary to ensure customer satisfaction.
For airlines, there’s an added opportunity for sustainability; the lighter an airplane, the less fuel required to get it where it’s going, which means both lower operational costs and a smaller environmental impact
So how can the transportation industry leverage artificial intelligence in supply chain management to counteract waste while securing customer satisfaction?
The Challenges of Handling Onboard Inventory
Let’s begin with the challenges faced by organizations planning for airlines, trains, cruises, ferries, and other businesses in the transportation industry.
Extensive, Complicated Networks
Transportation networks involve geographic factors, of course, but dealing with large – usually global – networks doesn’t just involve the physical distances. Planners have to plan across different regions, climates, and cultural expectations.
There is also the added complexity in terms of partnerships and parties involved.
Take airlines as an example. Many airlines have their own warehouses that feed “stations,” or stocking hubs, at the airports themselves. A catering company typically collaborates with these stations and handles the “last mile” aspect of loading the items onto the aircraft itself.
In other cases, such as cruise ships, the ship is the warehouse. Some ships restock at port, while others, in order to secure consistent product quality, carry everything passengers will need for the entire duration of the trip.
Regardless of restocking frequency and location, being able to collect and analyze data on traveler behavior, destination information, occupancy expectations, and trip duration is key to understanding demand for each item in your inventory. With demand insight of this caliber, companies can maximize customer satisfaction while reducing inventory costs and waste.
Tricky Product Portfolios
On top of network complexity, there is also the complexity stemming from the items themselves. For instance, different items are required for different flights depending on the trip duration and occupancy.
Plus, many of these items are perishable. Overstocking can lead to expiration and waste, causing real issues when it comes to sustainability, financial and environmental costs.
In fact, in its report “Sustainable Cabin: Cabin Waste and Single Use Plastics (SUP),” the International Air Transport Association (IATA) reports that 20-25% of cabin waste is untouched food and beverages.
With this in mind, responsible airlines and other companies in the travel industry understand that proper inventory management is about the environmental cost as it is the bottom line.
Data Disconnects: Inefficient Planning Tools and Techniques
Planners need systems that can handle copious amounts of data, deciphering signals, analyzing demand behaviors, weather, passenger demographics, when calculating proper inventory levels. However, many transit companies are hamstrung by outdated or inefficient planning practices.
Without an in-depth analysis of consumer behavior, planning teams end up stocking too much of the wrong thing and too little of the right thing, leading to waste, decreased customer satisfaction, and extra costs.
Today’s supply chains can no longer depend on spreadsheets and static planning tools to handle this current state of complexity. Even if the travel industry’s distribution network were less involved, there is still the ever-present risk of shortages or supply chain disruption.
Ready for Takeoff: How AI Is Laying the Foundation for Improved Supply Chain Practices for Onboard Inventory
With AI-driven dynamic planning solutions, companies have the planning horsepower they need to improve forecast accuracy, speed up decision making, and take decisive actions that reduce inventory investments while elevating passenger experience, protecting the environment, and improving profitability.
Rethinking The Supply Chain Framework and Supporting Best Practice with AI-Based Demand Modeling
Artificial intelligence is a wonderful tool, but in order to be applied correctly, one needs to consider the framework in which it operates.
This framework shifts from industry to industry and even company to company.
But reframing an industry’s standard practices can open new opportunities for applying best practices that have worked well in other industries, to new industries where they have never been employed. Let’s examine how.
Drawing Parallels: A CPG-Retail Allegory
In many ways, transit organizations mirror the CPG-retailer planning dynamic.
For instance, say you’re a massive CPG company, selling personal care goods. Looking at aggregated sales, you may be selling a lot of toothbrushes and floss and tubes of toothpaste.
However, at the item/location level (e.g., at the blue soft-bristled toothbrush/downtown Chicago level), you may only be selling a few of these items every couple of days. The aggregated view does not accurately reflect demand. It cannot provide the data analysis to represent granular demand behaviors.
Let’s apply this relationship model to the transit industry, with the airline warehouses and stations acting as the CPG company and the airplane being the retailer.
We find the same inherent challenge: a disconnect resulting from demand blind spots and poor supply chain visibility between two parties that share a mutual goal – high service levels, with low inventory. This challenge, involving two parties, makes this supply chain scenario a good candidate for Vendor Managed Inventory (VMI) techniques.
For example, an airline (our CPG here) may be going through a certain quantity of ginger ales. Ginger ale may seem like a safe stocking bet.
Looking at a granular level, however, perhaps only two people on that flight out of Logan wanted ginger ale. But the aircraft was stocked for far more.
If replicated throughout an airline’s supply chain network, how much extra weight, extra space, extra fuel, and extra money did this cost?
Considering that, according to that same IATA report, $2-3 billion in cabin resources ends up incinerated or in a landfill, the costs are quite significant.
Revolutionizing without Reinventing the Wheel
This is where demand modeling comes in, applying the same AI modeling techniques that help CPG companies and retailers collaborate and efficiently align inventory with demand.
By employing artificial intelligence, planners have the ability to forecast bottom-up instead of top-down.
This means that instead of aggregating sales history (or in this case, consumption history) and then averaging it across locations, an AI-powered modeling system takes a more granular look at data. It captures and decodes demand signals for each item, factoring in demand-related characteristics like the trip length, trip occupancy threshold, origin, destination, time of year, etc.
With this level granularity, the airline can see and account for the smaller ginger ale consumption on flights out of Logan in January.
And if this model were extended to encompass the different nodes of the airline’s supply chain, the suppliers at the airline’s warehouses and stations could achieve a symbiotic relationship with the catering companies that stock the onboard items, incorporating data from across the supply chain. This would increase supply chain visibility and reduce obsolescence stations and on the planes themselves, while still meeting passenger demand.
This reduction in stock would also protect the company from profit erosion and product expiration, as well as lessen the environmental impact from fuel consumption and inventory waste.
Transportation companies also need to account for the fact that many of the items served to passengers are loaded onto the plane by the tray or cart. This requires a firm understanding and control of the required bill of materials (BOM), best handled by a method called probabilistic BOM.
Probabilistic BOM allows planners to find the best possible array of component options when planning the carts and trays that are loading onto the aircraft or train
Probabilistic BOM accounts for the likelihood of certain components being used to determine the specific quantities of each component needed for the BOM, at the very start of the planning process.
As the BOM is used throughout the planning process, machine learning capabilities analyze and increase its accuracy, providing a clearer and more predictable view of component usage and protecting companies from over- or under-stocking each option.
This increases planning efficiency and aligns inventory with demand, decreasing the strain on the supply chain, regulating inventory levels for maximum working capital payoff, and boosting sustainability.
The Benefits of AI and Machine Learning
By harnessing artificial intelligence during the planning process, organizations in the travel and transportation sector are setting themselves up for improved business performance, higher customer satisfaction and a mountain of other benefits.
By reducing inventory, companies are likewise reducing the risk of obsolescence or expiration in the case of perishable items.
With the right dynamic planning tools, they can achieve this lower inventory while still having the right items in the right place to meet customer demand. This helps them maintain the level of onboard service that creates repeat customers and increases brand loyalty and reputation.
It also shrinks their carbon footprint, while improving working capital payoff and protecting margins.
In this way, AI drives the innovations that make supply chain a force for good – good for customers, good for businesses, and good for the planet.