Managing inventory effectively can be a challenge for healthcare and public sector supply chain teams. Balancing stock levels, reducing waste, and ensuring availability without overordering can feel endless. To further complicate the situation, many organizations rely on manual processes and gut instinct, which, as it sounds, often leads to inefficiencies and excess costs.
Enter Infor Inventory Intelligence: an intuitive solution within Infor CloudSuite that leverages data-driven recommendations to optimize inventory decisions. Curious about how it works? Good news!
On Friday, February 21, at 11:00 AM EST, Senior SCM Consultant Dorcas Spencer will walk through how to use Infor Inventory Intelligence to improve forecasting, automate supply replenishment, and enhance your supply chain operations. Key functionalities to be explored include recommendation engines, configuration settings, and Infor SCM integrations.
Join us as we break down how to set up and maximize Inventory Intelligence, run recommendations effectively, and engage key stakeholders to ensure long-term success. Session attendees will:
- Find out how Infor Inventory Intelligence makes managing inventory decisions easy
- Learn to configure recommendations and upload them with ISD
- Discover best practices for setup, security, and routine maintenance
- Gain expert insights on how to optimize inventory levels
Looking to level up your supply chain management this year? Check out the webinar and start managing inventory smarter, not harder.
Transcript
Dorcas Spencer
Hello. Today I’m excited to share with you about Inventory Intelligence. This tool comes with Infor FSM’s Inventory Control and is different from their newer AI Inventory subscription offering.
My name is Dorcas Spencer. I’m a supply chain consultant with RPI. What I’m going to cover today is a demo of Inventory Intelligence that will show some of the recommendations in our lightly used RPI sandbox, which should give you a picture — a flavor — of how this could benefit your organization. I’ll also review setup and how you can evaluate recommendations.
What is Inventory Intelligence? It looks at roughly 180 days to a year’s worth of transactions and makes recommendations on reorder points and max levels in your item locations. It analyzes transaction volume and reflects that into your reorder points and max levels, creating actionable information that you can take with a grain of salt as you evaluate your processes. You can even run it to apply updates.
What you need to know about your organization going in: Where are your heavy-hitting PARs and perpetuals? And what would be a good couple of PAR locations to start with to see the impact of Inventory Intelligence? You’ll want to get the right people involved — an IT resource to make sure replication runs properly, and supply chain leadership, inventory analysts, and coordinators to do the actual review. You can do your initial testing in your test environment or in production. It really comes down to timing.
The timing of when replication sets are running is where you’ll want to work with your IT folks. As the rules get set up, you’ll want to continue to evolve them — especially if you have a vendor that’s particular about which day of the week or when they’ll ship. You can create rules around those situations to help guide your evaluation.
Now I’m going to jump over to the application so we can look at some analysis. Here are the results of running the recommendation engine in the RPI sandbox. As you can see, we have limited activity. There are no increase candidates — nothing in here is recommended for an increase in quantity. There are several decrease candidates, shown here as the top ones, and then the removal candidates. On each of these tabs, you can click the arrow to get a list of the top items that need review. You can go back and do the same for the top removal candidates as well.
These three tiles give a summary. First, potential change: you can look at this $9.9 million reference inventory value and recognize there’s probably some action to take. In the sandbox, the recommendation is to remove $9.4 million. As you know, sandboxes have all kinds of activity going on — but not every day, and not all the time. The perpetual location tile tells you that about 56 of your larger items have some value you may want to keep, and then there’s a significant number you probably should remove. Your environment will have a little more involvement here. The PAR location tile shows similar data. The other two tiles are about history and do some analysis there. This is the Inventory Intelligence role view, and it gives you that high-level summary feel.
The next thing you want to do is look at your recommendations, and this view gives you much better detail. The first tab covers all locations. Then you have the perpetual inventory items and their details, and similarly the PAR items and PAR item details. There’s also an all-items tab that lists everything — very helpful when you want to look at one item and how it’s performing across the entire organization. You can see where it’s located and what recommendations have been made for it. For example, you might ask: why do they have 10 of this item and aren’t using them? That lets you evaluate that situation at an organizational level.
The filtering is really helpful for looking at actions and details one item at a time. When looking at recommendations, you can go into a PAR item and filter for those that have a recommendation. There’s a filter that lets you simply select Yes, click Search, and get that list. The same applies to perpetuals.
Here in the RPI sandbox, we have four items on one PAR that are flagged for review. You can see we’ve looked at the review process and decided we want to take action. On this item, I can choose to override or not override. I’m going to delete that and hit Save. I’ve received guidance from leadership that we’re not overriding for these three PARs we’re analyzing, so I won’t override. This one I’ve already marked as Approved and it won’t let me change that directly — I’d need to ship it back to Review, which would then allow me to modify the override.
On another item that’s already Approved, the next step would be to send it to the item location, which would update the item location record. Those are the main status functions available. Within each item, you can also right-click and put it out of scope. Out of scope removes it from consideration for this recommendation run and any future runs. Those settings can be changed at the system admin level. You can also Skip an item, which removes it from consideration for this run only. I use it as a way to note that I’ve looked at this item, don’t want any changes right now, and want to keep moving down my list. It helps with the review process.
Another useful feature: when you have a long list, it can be challenging to work through everything item by item. You can export the list to CSV — I’ll switch it to Excel because I know I’m going to want to do an upload from it. Here is that same recommendation list. I’ll enable editing and run an upload.
This allows you to select the FSM business class, which is IIH Item Location. From there, you have options for a variety of actions. What’s great about this is it supports every step of the process. In this case, I’m going to move items from Review to Approve. Review is the current state, and I want to switch it to Approve. I’ll click that, click Insert, and then run the upload for just that one line and hit OK. As I scroll over, you can see it says Approve is complete. I’ll refresh the screen and you’ll see it changed from Review to Approve. This is a very efficient way to process a large number of items at once — you can move from Review to Skip, Review to Approve, and so on. It gives you a lot of flexibility in managing that workflow.
Within this recommendation view, there’s a lot of information provided. You can enable the search option and do some filtering on your data. For example, if you want to focus on items with a dollar impact over a certain threshold, you can filter accordingly. Those filters help you focus on what’s most important.
That’s how recommendations can be reviewed within the system. It’s very handy. I’ve also added Company and Inventory Location as a personalization here, which helps when doing exports.
For rules, you can create rules in the Inventory Intelligence Specialist role. These are not the complete set of rules — we’ll talk about that in a moment. But those rules can be set up to evaluate various conditions. Going back to the presentation: the Generate Daily Data function creates a chart of activity, which is pretty interesting — you can see where you’ve had to restock frequently or not very much for your items.
The setup is pretty straightforward. You go to Global Configuration in the Supply Management setup, follow the Inventory Intelligence configuration path, and once you’ve created that global configuration, you get the option to set up replication. You click that, select the data area, and click Submit. That starts the replication, which fills up the replication dataset. Once that’s complete, the engines run the daily data to create those graphs, and then the recommendation engine runs. There’s a results screen so you can track whether processing is in progress or complete.
There are two security roles to keep in mind. The Inventory Intelligence Specialist has access to a specific set of rules. The Supply Management Administrator has access to global rules, location groups — which is a useful feature that lets you run the engine against a specific group of locations — and access rights management. These two roles are important for the initial setup, with the administrator role being especially important for ongoing maintenance.
Here’s your global configuration. This one has a replication set, and you can see when it was last run and kick off a new replication pull from here. All the Inventory Intelligence processors are set up here, similar to a Requesting Locator setup. Access is either Edit, View, or No Access. Typically an organization might set things up so that everyone can view, but users can only edit the locations they manage.
Within the global configurations, these four rules are the overall rules that the Inventory Intelligence Specialist role cannot edit — only the administrator can. A quick overview of each: Data Cleansing recommends at least 180 days of data. Data Filtering addresses things like unit-of-measure changes — if you had a change, you may want to exclude prior data to avoid a mismatch in analysis. Quantity upper and lower bounds handle outliers — if a million units passed through, the system can exclude that from the analysis set so it doesn’t skew your results. Lead Time has defaults and configurable minimums to help with how lead time factors into the analysis.
Post Processing rules help prevent minor or insignificant recommendations — they ensure recommendations are meaningful rather than something like a penny here or a penny there. Both decreases and increases have thresholds, so for example, the system only makes a recommendation if the change exceeds a defined minimum. As you review your data, you may want to tweak these. Perpetual global parameters include settings around how much history to use and how many days to include in a simulation. PAR global similarly addresses transaction history and PAR impact by unit of measure. All four of these global rules are very helpful as the engine analyzes data for new items, or items whose usage is slowing down.
I’ll wrap up with some thoughts on putting this into operations. Run the daily engine every day. For the recommendation engine, I’d recommend running it quarterly, or perhaps every other month, because it will overwrite the previous recommendations. If your team is actively working through recommendations and timing when they send updates to item locations, you don’t want a new run overwriting work that’s in progress.
Encourage participation. Sit down with department managers on their PARs as you review what should be added, what should be reduced, and what space they have available. Highly recommend reviewing those recommendations with filters like the dollar savings view I showed you, and starting with a few locations as a first pass — maybe focus on two PAR locations and a specific group of items in inventory.
After you’ve gone through a first or second pass on the recommendation engine, start thinking about where you could modify the global rules — data cleansing, post processing — and the specific rules that can tie to an item, a location, or a vendor. Make it engaging. You’re going to have some PARs that look spot-on, and a handful where you’ll want to make some tweaks. I recommend getting in there, running the data replication, running the recommendation engines, and enjoying the opportunity to analyze your reorder points and max quantities. Thank you for your time. Have a great day.
