Supply Chain Optimization is to Supply Chain Managers what flight simulators and autopilot are to pilots. To expand, the Digital Twins generated are the flight simulator of a company's Supply Chain Networks that enable Supply Chain Managers to test different Replenishment and Routing Strategies in a virtual model. Once the optimal strategy is calculated, the optimiser acts as an autopilot: autonomously executing the optimal business in the real world to allow you to efficiently achieve your goals.
Supply Chain Optimization is a key component of developing a competitive supply chain strategy. It leverages two key technologies, namely Digital Twins and AI Optimization, to enable leading Retailers to uncover inefficiencies in their Supply Chain and develop better Inventory Policies that align with their business goals.
It is accomplished by replicating your Supply Chain in a digital twin, and reconstructing historical Supply Plans or Routing Choices. Once the digital twin is setup and validated, an AI Optimization is setup to test different strategies in the Digital Twin. The AI Optimization automatically and autonomously identifies the best strategy that simultaneously maximises your Customer Satisfaction while minimizing Operative Costs.
These strategies are run through backtesting in the Digital Twin by the AI Optimization, and the AI Optimization will identify the Inventory Policies that best meet the objectives chosen by the Supply Chain Manager. Then the results and statistics from the backtesting are presented to the Supply Chain Manager to validate the effectiveness of the strategy.
The underlying theory of Supply Chain Optimization and backtesting is that any Replenishment strategy that worked well in the past is likely to work well in the future, and conversely, any strategy that performed poorly in the past is likely to perform poorly in the future. This article takes a look at what use cases are available in Supply Chain Optimization, what kind of data is obtained and how to put it to use.
In Retail, Optimization is the process of fine-tuning a Replenishment Strategy or Routing Strategy in order to improve efficiency or reduce costs. This can be done by using machine learning to automatically find the best parameters in the Supply Planning Process to align with the businesses priorities. Enabling retailers to run their replenishment or routing more efficiently. For example an Optimization can be run to find the best Days Cover Settings, by SKU & Location, to maximise Fill Rate while keep Inventory Holdings below a specified threshold.
It is worth noting though that most optimizations come with trade-offs and opportunity costs in other areas. For example, a retailer that optimizes to reduce its stockouts is also likely to increase total stock held, and a company that optimizes by reducing labor costs might find itself short-staffed in the event of a sudden increase in demand. This is why it’s important that the Supply Chain Optimization Solution you use has backtesting capabilities and detailed reporting that breaks down the opportunity costs and overall impact of the new strategy.
There are a number of use cases for Supply Chain Optimization. Including:
Leverage the AI Optimization and Backtesting to find the best Supply Planning policies. Whether it's a Min-Max, Material Requirements Planning (MRP), or Days Cover policy, the best Supply Chain Optimization Softwares enable you to automatically find the best policies, and rapidly test their impact on your Supply Chain, both in Historical Backtests or in Simulated Scenarios. When evaluating the new Planning Policies, you can compare them against your existing policies and stress test them under different conditions, such as increases in demand, supply disruptions, new store openings. This allows you to rapidly and holistically see how the new Planning Policies will work in your Supply Chain Network.
Leverage the AI Optimization to find the routing inefficiencies in your Network, identify which specific skus should be ranged in different DCs to get them closer to the final retail outlet and reduce freight and Co2 output across your network by reducing the miles/kms your products travel to get to the retail outlet.
Leverage the Simulation to model the complex interdependencies of the Global Supply Chain, optimizing the Inventory Policies across the network, accounting for Supply in all aspects of the Network.
In addition to optimizing inventory policies across the network, An end-to-end supply chain simulation can act as a Supply Chain Control Tower, providing the required visibility into multiple echelon operations.
Typically, Optimization Software will have two important screens. The first allows the Supply Planner to customize the settings for testing. These Settings include everything, from Product Selection to Time Period. Here is an example screenshot from the Settings Screen in Remi AI:
The second screenshot shows you the improvement on Lost Sales by month in the backtesting results report. It shows that there is 57% reduction in lost sales to be found with different replenishment settings. Again, here is an example of this screen in Remi AI:
Objectives are a critical concept in Supply Chain Optimization. They are the goals of what you’re trying to optimize. They're typically either something the Retailer is trying to Maximize or Minimize and they're typically defined at a total business level, such as Minimizing the Lost Sales across the Supply Chain. Some examples of common objectives include:
- Minimizing Stockouts
- Minimizing Inventory Holding Costs
- Minimizing Freight Costs
- Maximizing Availability/ Fill Rate
- Minimizing Lost Sales
- Minimizing Days Cover
The constraints define the rules of a legitimate solution in the Digital Twin (and in the Real World). For example, if you want to minimise costs, it is probably best to not make any products, not ship anything, or have no facilities. A cost of zero is indeed the minimum; however this is clearly not realistic. So there are some logical constraints we must include in the optimization if we want to meet all the demand.
Constraints are used by the Supply Chain Manager to help control the Optimization and ensure it delivers policies that are aligned with the Business Objectives.
Some examples of common constraints include:
- Keep total Inventory Costs below $X
- Keep Days Cover below X Days
- Keep Service Level above X percentage for A Class SKUs
Backtesting involves testing an approach or inventory policy in a historical simulation to study it performance. It is an approach commonly used by traders and hedge funds to check the viability of a trading strategy and the leading Supply Chain Optimization Softwares, such as Remi AI, include advanced backtesting tools that give Supply Chain Managers similar capabilities to the leading hedge funds. It is important to note that while it is not guaranteed that an Inventory Policy that performed optimally in backtests will continue to do so well into the future, it is still the most advanced and comprehensive way of finding the optimal policies.
Direct Cost Savings
The strongest argument is always a direct reduction in expenses. Supply Chain Optimization is the best practiLess truck miles to serve the same number of stores, fewer containers with the same amount of goods. This directly boosts the bottom line.
Less capital cost
New managers often don’t understand the concept of tied-up capital until they receive a vendor invoice before being able to sell the product. “He wants $5000 right now? I don’t have that money!” Well you do - it is sitting right there on your warehouse shelves. Too bad that it isn’t cash, and you really have to make a sale before you can free it up. If your warehouse is skinny and your supply chain is lean, you will have much less of that elusive working capital tied in your warehouses and transport networks.
For the next ten years or so, everybody will remember the Covid-19 pandemic and how it disrupted supply chains. But it doesn’t have to be a global disaster, smaller events can also cause severe disruption. Digital Twins enable Supply Managers to test entirely new scenarios, such as unprecedented supply gaps, new retail outlets, new Demand. So just how Pilots can practice emergency procedures from the safety of their flight simulators, Supply Chain Managers stress test inventory policies under custom simulated scenarios,
This enables them to build the resilience of their Supply Chain resilience.
A simulation is often meant to play through a scenario in theory, and to gain insight into what might happen. A digital twin of a supply chain allows you to narrow down generic predictions to quantitative forecasts. Instead of just knowing that a warehouse in Tonawanda will reduce stockouts for two important market, you can actually determine that 8,000 sqft will be the ideal size. Or that daily truck runs directly to stores will be cheaper than a regional center with delivery vans.
In summary, optimizing your replenishment and routing operations means more efficient product replenishment, more profit, and a more efficient supply chain. It is a best practice to be using this technology if you’re managing a range of more than 5,000 skus.
Replenishment Strategy is definitely an area where many retailers’ operations are not following best practices. While a single retailer may not be able to employ all best practices, they should prioritize the most feasible and impactful development areas from the business’s perspective.
Supply Chain Optimization is driving real change for a lot of retailers: A good replenishment strategy can identify significant opportunities across your Supply Chain quickly, and more cost-effectively that a fine-tuned team of analysts. Liberate your Supply Planners and Managers from the depths of data analysis so they can actually use their expertise, making the critical decisions that a computer can't and making your customers love you.