Back in late 2018 and early 2019, before the GENizah Artificial Idiocy craze began, the doctor did a sequence of AI Series (totalling 22 articles) on Spend Matters on AI in X Today, Tomorrow, and The Day After Tomorrow for Procurement, Sourcing, Sourcing Optimization, Supplier Discovery, and Supplier Management. All of which was implemented, about to be implemented, capable of being implemented, and most definitely not doable with, Gen-AI.

To make it abundantly clear that you don’t need Gen-AI for any advanced back-office (fin)tech, and that, in fact, you should never even consider it for advanced tech in these categories (because it cannot reason, cannot guarantee consistency, and confidence on the quality of its outputs can’t even measured), we’re going to talk about all the advanced features enabled by Assisted and Augmented Intelligence that are (or soon will be) in development (now) and you will see in leading best of breed platforms over the next few years.

Unlike prior series, we’re identifying the sound, ML/AI technologies that are, or can, be used to implement the advanced capabilities that are currently emerging, and will soon be found, in Source to Pay technologies that are truly AI-enhanced. (Which, FYI, may not match one-to-one with what the doctor chronicled five years ago because, like time, tech marches on.)

Today we continue with AI-Enhanced Supplier Management that is in development “today” (and expected to be in development by now when the first series was penned five years ago) and will soon be a staple in best of breed platforms. (This article sort of corresponds with AI in Supplier Management The Day After Tomorrow that was published in May, 2019 on Spend Matters.)

TOMORROW

Supplier Future State Predictions

Supplier management platforms of today can integrate market intelligence with community intelligence, internal data, and external data sources and give you a great insight into a supplier’s current state from a holistic perspective.

Along each dimension, future states can be predicted based on trends. But single trends don’t tell the whole story. Now that we have decades of data on a huge number of companies available on the internet across financial, sustainability, workforce, production, and other dimensions which can be analyzed overtime and cross-correlated, we can do more, and know more.

Based on this correlated data, machine learning can be used to build functions by industry and company size that can predict future state with high confidence based upon the presence of a sufficient number of sufficiently accurate data points for a company in question. Now that these platforms can monitor enough internal, community, and market data and pull in a plethora of data feeds, they can accurately compute metrics with high confidence along a host of dimension, and this in turn allows them to compute the metrics that are needed to predict future state if the vendor’s platform has enough historical data on enough companies to define trends and define predictor functions using machine learning.

Not only can you enter a relationship based on a current risk profile, but on a likely future risk profile based on what the company could look like at the end of the desired contract term. If you want a five year relationship, maybe taking advantage of that great deal due to a temporary blip in supplier or market performance may not be a good idea if suppliers historically in this situation typically went into a downward spiral after accepting a big contract they ultimately weren’t prepared to deliver on.

Category Based Supplier Rebalancing

We could actually do this today, as a few vendors are now offering this capability, but it’s not yet part of supplier management platforms and the newly emergent offerings are often limited to a few categories today. But tomorrow’s platforms will continually analyze your categories holistically (along the most relevant dimensions, which could include cost, supply assurance, environmental friendliness, etc.) to determine if the supply mix you are currently using is the best one, let you know if there could be a better one, and suggest changes to orders (as long as it doesn’t jeopardize contracts where that jeopardy could come with a financial or legal penalty).

It’s just a matter of re-running an optimization model on, say, a monthly basis with updated data on price, supply assurance, and environmental friendliness (using the appropriate data for each, such as market quotes, current supplier risk, carbon per unit, etc), and comparing the optimal result to the current allocation plan. If it’s within tolerance, stay on track; if it’s slightly out of tolerance, notify a human to conduct and review a thorough analysis to see if something might need to change; if it’s way off of tolerance, recommend a change with the data that supports the change.

Supply Base Rebalancing

Once you have a platform that is continually reanalyzing categories and supplier-based assignment, you can start looking across the supply base and identify suppliers which are hardly used (and an overall drain on your company when you consider the costs of maintaining a relationship and even maintaining the supplier profile) and supplier that are potentially overused (and pose a risk to your business simply based on the level of supply [as even the biggest company can stumble, fall, and crash to the ground on a single unexpected event, such as the unexpected installation of a spreadsheet driven Master of Business Annihilation as CEO who has no clue what the business does or how to run it effectively and, thus, causes a major stumble, as summarized in Jason Premo’s article).

And, more importantly, identify new suppliers who have been performing great with slowly increasing product / service loads and should be awarded more of the business over older suppliers that are becoming less innovative and more risky to the operation at large. Now, this will just be from a supply perspective, and not a supply chain perspective (as these programs focus on suppliers and not logistics or warehousing or overall global supply issues), but this will be very valuable information for Sourcing and New Product Development who want to always find the best suppliers for a new product or service requirement.

Real-Time Order Rebalancing

Since tomorrow’s platforms will be able to recommend category rebalancing across suppliers, they will also be able to quickly recommend real-time order rebalancing strategies if a primary supplier is predicted to be late in a delivery (or a human indicates an ETA for a shipment has been delayed by 60 days). This is because they will be integrated with current contracts, e-procurement systems, and have a bevy of data on projected availability and real historical performance. Thus, it will be relatively simple to recommend the best alternatives by simply re-running the machine learning and optimization models with the problematic supplier taken out of the picture.

Carbon-Based Rebalancing

Similarly, with the rise of carbon-calculators and third-party public sources on average carbon production per plant, and even unit of a product, it will be relatively easy for these supplier management platforms to build up carbon profiles per supplier, the amount of that carbon the company is responsible for, how those profiles compare to other profiles, and what the primary reasons for the differentiation are.

The company can then focus on suppliers using, or moving to, more environmentally friendly production methods, optimize logistics networks, and proactive rebalancing of awards among supplier plants to make sure the plants producing a product are the ones closest to where the product will be shipped and consumed. It’s simply a carbon focussed model vs. a price focussed one.

SUMMARY

Now, we realize some of these descriptions are dense, but that’s because our primary goal is to demonstrate that one can use the more advanced ML technologies that already exist, harmonized with market and corporate data, to create even smarter Supplier Management applications than most people (and last generation suites) realize, without any need (or use) for Gen-AI. More importantly, the organization will be able to rely on these applications to reduce time, tactical data processing, spend, and risk while increasing overall organizational and supplier performance 100% of the time, as the platform will never take an action or make a recommendation that doesn’t conform to the parameters and restrictions placed upon it. It just requires smart vendors who hire very smart people who use their human intelligence (HI!) to full potential to create brilliant Supplier Management applications that buyers can rely on with confidence no matter what category or organization size, always knowing that the application will know when a human has to be involved, and why!