How AI is Creating Greater Influence and Visibility for Retail Supply Chain Planning
Do your company’s supply chain leaders have an influential role in shaping your brand’s decisions?
If your organization’s supply chain leaders haven’t gained prominence in the wake of the myriad supply chain disruptions over the past few years, the proliferation of artificial intelligence technologies might be what boosts their visibility among the C-suite and board of directors.
AI-enabled retail supply chain planning technologies have the potential to streamline and automate rote aspects of collecting, cleansing and harmonizing data, which allows planning teams to focus on more analytic and insights-driven aspects of their roles. Additionally, AI can help analyze a company’s historical data, along with relevant external data points, to help create forecasts with improved accuracy that can influence purchasing and promotional decisions for the business.
Here are some of the ways that AI is helping to elevate supply chain planning functions and capabilities.
Bringing Strategic Thinking to the Forefront
AI tools can transform the daily tasks of retail planners by allowing them to focus less on tactical aspects and focus on more strategic aspects of their role.
Let's take an example like compiling data for integrated business planning (IBP) reports. Traditionally, the work required to prepare an IBP report could take hours or even days to account for the segmentation of data, extracting and merging into Excel, the replacement of null values, insertion of non-master data, and correction of calculations, the porting into PowerPoint, the re-sizing and updating of slides.
The advent of AI allows planners to pass the creation of IBP reports to the solution: the planning manager can simply request the AI tool to build the required material. Legacy planning solutions might pre-define an IBP report for creation on request, but AI can research and create material based on ad-hoc descriptive text. Think of ChatGPT's capability within your planning system.
Automating Risk Management and Opportunity Leverage
AI platforms allow planners to plug in their parameters. Then the platform’s AI algorithms can pull the data necessary for the planner to focus on exception management and determine relevant insights from the dataset to make better decisions across the business and evaluate potential risks and opportunities. For example, if a competitor is facing an inventory shortage, your planners could determine which expediting scenarios could position a competitive advantage and increase sales. Similarly, given the appropriate targets, an AI/machine learning solution can recommend the best promotion to boost sales or re-route a distribution plan from the most cost effective to the fastest delivery.
Creating More Accurate Forecasts
Historically speaking, demand forecasts were often inaccurate. As little as five years ago, demand forecasts could be wrong by as much as 50 percent. As a result, leaders didn’t put much faith in the results or factor them into business decisions. But AI has changed the game on forecast accuracy. Companies with irregular data using historical sales activity and time series forecasting will continue to see unimpressive results as the time series methods cannot easily adjust for changing demand drivers, whereas ML forecasting using multiple variables can recognize and react to changing demand patterns quickly. AI allows planners to gain insights from historical promotion and sales data and current external drivers (e.g., weather, upcoming events and holidays, etc.) and better predict what demand will look like in the near future.
An Evolving Statistical Base and De-Risking Planner Insight
Another beneficial aspect of AI/ML over time series is deep learning. There are two significant impacts: The first is system knowledge. With every forecast cycle, the base statistical forecast will continue to learn and improve, whereas a time series solution will only deliver results in line with its source data. Accuracy improvements are baked into the design of an AI/ML solution. The second is planner knowledge. With legacy systems, a demand planner leaving an organization will take their insight with them and the new planners will need to learn afresh. With AI/ML, newly onboarded demand planners aren't enriching a baseline from scratch because the baseline is already enriched. New planners are able to get up to speed faster and with less risk to the organization.
Bringing it to the Board
As AI algorithms continue to offer insights from the data and improve processes and decision making, retail planners are able to deliver consensus forecasts more frequently (weekly instead of monthly, daily instead of weekly). Faster forecast cycles allow organizations to see how demand for specific products is shaping up, which factors are influencing product consumption, and if a company needs to make pivots to its business plans to capitalize quickly on emerging opportunities. Supply chain leaders who are able to make the business case on the possibilities that AI can bring to their business and supply chain and how it can be beneficial to the company’s overall growth will likely secure more influence at the executive leadership level.
Simon Joiner is a director of product management at o9 Solutions, the knowledge powered analytics, planning and learning platform for next-generation global enterprises.
Related story: How Real-Time, Item-Level Inventory Visibility Drives Retail to Efficiency