Are We Building Fragile Supply Networks?

This article was originally published in the Supply Chain Alpha newsletter on LinkedIn. Read the original here.

"The resilient resists shocks and stays the same; the antifragile gets better."

— Nassim Taleb in "Antifragile"

The Truth About Outsourced Manufacturing

A few decades back the shift to China ushered in the age of contract manufacturing. Regardless of the direction of current geopolitical forces and the trend away from China, the concept of outsourced manufacturing is here to stay. And to be clear, it has proven to be a highly effective ecosystem strategy to develop manufacturing expertise at scale among a few outsourced providers while brands focus on the product design and market. However, as we scaled supply chains from these outsourced factories, we also outsourced the supply chains behind them. We are talking about those countless suppliers and components that find their way into our products. Today, outside of a few exceptions, the entire planning and orchestration of this materials supply chain is managed by the contract manufacturer - not the Brand. Sure, most brands will play an active role in managing critical component sourcing and supplier relationships - however, the tactical planning, monitoring and procurement of components is outsourced. And this is precisely where the risk hides and builds up, like seismic fault plates building tension, only to break down seemingly randomly in the form of a supply chain disruption. In this edition, I discuss why we suffer impacts from these "hidden" risks, what incentives are at play and how to get around to exposing and mitigating them.

The Visibility Challenge

The word "Visibility" comes with connotations. The industry has software vendors with visibility solutions extending to many tiers of suppliers. But this is "visibility" with an intent to establish knowledge about the network's footprint - undoubtedly useful to know the exposure and blast radius reactively in the event of a disruption. But there is another kind of visibility much closer to home that most brands tend to ignore. The kind that directly tells you if you have the right supply of the right materials at the right time to meet your production and in turn, your S&OP plans and your customer commitments.

But first, lets review the basics of an outsourced manufacturing set-up. Brands provide their factories with a supply or build plan and issue POs against it within an agreed upon time horizon. In return, the contract manufacturer (or CM), procures all components to support the build plan and ships POs on time and in full. The CM takes on the entire burden of planning all commodity components while the brand may choose to own planning and procurement for a few critical suppliers where they have higher leverage on cost, capacity, quality, technological investments etc than the CM does. But for all intents and purposes, the CM and its team runs the Materials Requirement Plan (MRP), and manages/monitors all tactical component scheduling.

It is in this outsourcing of the management of the materials supply chain that we find the greatest opportunities to build resilience.

To start with, MRPs are legacy engines, highly complex and custom installations with limited to no ability to integrate easily with modern data platforms and applications. Moreover, given their low margin business model, CMs employ junior MRP analysts to interface with their customers' supply chain teams to review materials supply. These analysts publish "Continuity of Supply" (CoS) and "Clear to Build" reports, typically in the form of weekly excel files comprising millions of rows of component level data. These reports are cumbersome to analyze, contain limited context or master data and lack exception flagging to prioritize response. Companies have largely built tribal knowledge processes to handle these conversations with the CM. However, any notion of proactive planning and risk mitigation is sadly absent. Structurally, the CM is incentivized to keep an arm's length model with their customers due to shared assets and as such, even tactical tools like MRP are not directly exposed to their customers' supply chain.

This opacity in the materials planning process, in my opinion, has severely limited the ability of supply chains to sense and respond to the build-up of risks. In addition, the inability to audit, modify and leverage MRP master data (think lead times, MOQs, alternate supplier volume allocations etc) adds a systemic risk to the flow of materials. I believe if these structural blockers are removed, supply chains would double their speed to detect and respond to risks, increase topline attainment (+1-5%) and improve the bottomline (+50-100 bps).

Anti-fragility could also be emergent as the supply chain gets better at detecting and capitalizing on risks/volatility and thereby, winning market share from competition while retaining or even improving margins.

Now the key question is "How"? The answer is fairly straightforward…

Resilience in the Materials Supply Chain

The first step to building awareness of risks brewing in your supply network is to digitally ingest, analyze and harvest insights from the CoS and Clear-To-Build data that is already shared by every single CM with each brand. This creates a digital repository of materials procurement status across open orders, forecasts, inventory and production plans phased over time. Investing in a data layer for data harmonization, cleaning and integration is crucial here. Further, integrating the brand's bill of materials (BOM) and Supplier data from Product Lifecycle Management (PLM) and Supplier Relationship Management (SRM) systems is crucial. You are now ready for the next steps to systematically harvest this repository for risk insights and actions.

A key challenge that technology helps overcome at this stage is parsing through millions of records of component level data to rank and prioritize the most impactful supply shortages and excess inventory risks for resolution. Deploying machine learning (ML) based scoring of risks based on degree of impact is a great use case that can be rapidly launched. The key benefits of monitoring this data includes:

Proactively flag shortages outside lead time (at least 2x Faster)
Design a workflow solution between Supply Chain, Sourcing and CM teams; for example a ticket or case-based system to track shortage/excess inventory scenarios
Track and flag out-of-threshold demand, supply and master data movement
Use AI for predictive risk identification and simulation modeling
Feedback all of the above into the S&OP for smarter allocation of finite supply for profitable demand responses
Hold CM and Suppliers accountable to On-Time and In-Full performance against forecasts and POs

The above capability cuts down the time from data to insight to action by at least 2X. Team productivity dramatically improves as they spend less time triaging across excel/email/meetings. Instead, they are directly focused on the highest priority risks with all supporting data and recommendations to resolve (as AI predictability improves over time). It is important to ensure that the AI engine is trained on specific data relevant to the supply network instead of broadly pointing to all enterprise data.

Resilience as a capability is best built with the mental model of a flywheel. With increasing data, the pattern recognition capabilities improve and prioritized workflows enable teams to respond with increasing speed, accuracy and effectiveness in mitigating risk.

With more maturity, teams can move beyond mere reactive mechanisms to proactively enacting strategies for resilience including -

Lead time reduction investments with critical suppliers
Buffer investments where the risk-adjusted return on investment is positive
Investments in Capacity Redundancy in key nodes
Negotiating pricing and allocation mechanisms where applicable with suppliers

Over time, it is reasonable to expect this paradigm could extend into deeper tiers of the supply chain.

Challenges with this approach

As straightforward as it is, even this approach has challenges.

First, it is unclear if there is a sense of urgency in investing in the technology and resources to build this capability. Many companies expect their CMs to take care of the materials supply chain and only expect to be involved for crisis resolution. The common refrain is "Isn't that why we pay our CMs?". Supply Chains need to understand that outsourcing manufacturing does not mean outsourcing your ownership around managing end-to-end risks in your ecosystem.

Second, the ownership of the relationship with the CM is complex with Sourcing, Procurement, and Supply Chain all involved but a single owner who has the organizational authority to sponsor such a capability is rare. This may well come down to organizational structure and ultimately, who is most accountable for ensuring supply continuity to meet committed S&OP plans. Ideally, the Chief Supply Chain Officer with ownership across Planning, Procurement and Sourcing is the right stakeholder to push for this solution.

Lastly, with multiple CMs with different reporting formats, the standardization of the data platform is a non-trivial exercise that needs sponsorship. To get around this, a company needs to have a consistent interface with each CM, standardized data ingestion and harmonization protocols and focused integration efforts with each CM. Unfortunately, CMs do not employ IT resources for free and these integrations come with one-time and potentially, small ongoing costs.

In Conclusion

Supply resilience does not necessarily need complex n-tier supplier integration to get started. Far more practical and impactful resilience can be built by harnessing the information between a company and its manufacturing node. With a strong data integration platform, the use of AI-based risk detection and scoring allows humans to focus mitigation efforts on the biggest risks where failure could be damaging. As always, building this capability needs clear ownership and investment from senior supply chain leaders. As global trade networks are reconfigured with tariffs and geopolitical dynamics, supply chains that build this capability with their manufacturing partners will take disproportionate winnings in the market.