by demo3 | Jan 16, 2026 | Homepage Featured
As trade agreements shift, tariffs fluctuate, and geopolitical tensions escalate, manufacturers need to be more strategic and adaptable than ever before. Old assumptions about cost, scale, and stability no longer hold in today’s increasingly fragmented global trade environment, and there is mounting pressure for manufacturers to rethink how and where they produce goods.
However, there are practical strategies manufacturers can use to improve the flexibility and resilience of their operations by designing supply chains and production networks that can adapt quickly to sudden changes. The organizations that come out on top in this environment will be those that can identify vulnerabilities in current configurations and rapidly adjust operations to reduce risk and manage costs.
Whether you’re preparing for the next disruption or aiming to manage long-term operational risks, it’s important to understand the factors impacting the current global trade landscape, what’s likely to come next, and what actions you can take to stay competitive in unpredictable conditions.
How did we get here? The emergence and future of a volatile global trade environment
After World War II, a multilateral trading system evolved, culminating in the World Trade Organization (WTO) and supplementary regional trading blocs. In this trading regime, stability was the norm, with consistent, negotiated arrangements punctuated by occasional renegotiations. Over this period, the system became not only more stable but generally more open, with lower and fewer tariffs, albeit with increasingly elaborate non-tariff barriers (NTBs). This system relied on the United States to remain largely open to imports, absorbing manufactured exports from most other Organization for Economic Co-operation and Development (OECD) countries, while exporting services.
https://www.census.gov/foreign-trade/statistics/historical/index.html
Recently, the new U.S. administration made significant changes to this trading regime. This was only the final step for a trade ecosystem that had been eroding over time. There have been many signs of this breakdown:
- Resilience, self-reliance, and national security have been overtaking efficiency, scale, and low-cost production as national goals. After COVID-19 exposed vulnerabilities, China, the EU, and the U.S. are all aiming to near-shore, reshore, and “localize” key production.
- Trade and technology have been weaponized (e.g., export bans), and industrial policy, with its attendant protectionism, has spread — even to the U.S.
- Regional blocs have emerged, including the United States-Mexico-Canada Agreement (USMCA); the Belt and Road Initiative (BRI) and the Digital Silk Road (DSR); India, ASEAN and the EU; and South-South trade.
- The erosion of WTO authority and associated multilateral norms.
There are reasons to believe this is a permanent rupture and not a shock that can be recovered from. The forceful abrogation of the WTO, USMCA, and other multilateral (and bilateral) trade agreements by the U.S. (followed by most other major traders) will make it very difficult to put the previous state back together again. Even if new multilateral agreements could be made, why would any partners believe the U.S. would honor them? Trust must first be rebuilt through sustained, reciprocated trade actions.
In addition, important conditions attending the development and maintenance of the WTO regime no longer hold: American economic dominance; American leadership and support for multilateral institutions supporting open trade (the General Agreement on Tariffs and Trade (GATT) and then the WTO, the International Monetary Fund (IMF), the World Bank); and Cold War geopolitics that encouraged the U.S. to support open trade with allies. Without these conditions, it seems unlikely that the global trade environment would return to its prior state.
We can’t know the details of whatever new trading regime evolves, but a few characteristics seem likely:
- More trade through bilateral agreements and less through multilateral agreements (e.g., WTO rules only), as bilateral agreements are easier to reach and to modify than multilateral agreements among several or many countries.
- Multilateral agreements that persist will be more regional and include fewer countries.
- NTBs will be complementary to tariffs — lower stated tariffs will be offset by more NTBs (and vice versa).
The result is likely to be a patchwork of shifting trade agreements — with tariffs; NTBs (such as rules of origin, labor and environmental standards, quotas, subsidies, and certification requirements); and exchange rate regimes and currency policies all changing more and more frequently than was the case in the trading regime just ending.
The effects will be felt in the prices and availability of both manufacturing inputs and finished goods at their destinations.
Configuring production and supply chains for resilience
Given these uncertainties, there is a lot of concern about supply chain resilience. Often, supply chain resilience is described as the ability to withstand disruptions and commonly measured across three dimensions:
- Time to Recover (TTR)
- On Time In Full delivery rate (OTIF)
- Order cycle time
But what does resilience mean in the context of shifting trade rules? There is generally some persistence to changes in tariffs, NTBs, or currency regimes — at least long enough to make some adjustments in sourcing or what gets produced at each location. While this has not been true for recent disruptions, that is an exception and not the norm. Typically, changes stick around long enough to offer leaders opportunities to shift strategies and compensate. A more comprehensive, strategic perspective is needed to navigate this new environment:
Viewing your set of production facilities and suppliers as a configuration, resilience means how well that configuration can adapt to step changes in tariffs, NTBs, or currency regimes. One way to measure that is in total production cost (the cost to manufacture the full set of products to meet demand for a quarter or a year). How much does total production cost change when particular tariffs change, or exchange rates shift, or export controls spike the price of a key input for months or years?
Building a profile of how the total production cost changes in response to plausible scenarios can tell you how vulnerable you are to such shifts. And if you are configured such that you can shift (some) supply inputs to countries unaffected by changing trade rules, or move production of (some) finished goods to factories similarly un- (or less) affected, you can reduce the impact of the trade rules change on total production cost.
The most resilient production configurations will have some overlapping manufacturing capabilities and be dispersed enough to reduce the impact of repeated, unpredictable trade rules shifts.
Supply chain and production resilience is not costless — you must build some redundant manufacturing capacity, maintain a distributed portfolio of key suppliers, and have the ability in your management systems to adjust sourcing and S&OP to adapt to changes in trading conditions.
The only way to know if this cost is worthwhile is to model your current configuration of suppliers, factories, and sales destinations and test it against changes in tariffs, NTBs, and exchange rates. Within your current capabilities, can you make changes (in sourcing or in what is produced at each factory) to reduce the impact of trade shocks? Are there changes that can be made to your configuration (e.g., adding a production line at a particular factory or adding a supplier from a different region) that would materially improve your resilience? Modeling and optimization can point you to the answers .
For example, a F50 manufacturer of large durable goods with a sprawling global production base wanted to optimize “what gets made where” in order to reduce costs while improving flexibility and increasing their resilience to shocks in trading conditions. We worked closely with executives and an operational core team to:
- Establish, define, and measure the key parameters to balance, including: cost, quality, risk and flexibility
- Determine and acquire key inputs such as product data, labor costs, tariffs, exchange rates, shipping rates, production costs, component/raw material supply costs, and many more
- Develop target scenarios and iteratively model both the financial and non-financial impacts of each scenario (comprehensive multi-variant optimization model)
- Assess results and recommend “optimal” global manufacturing network – including dramatic changes in product modularity and manufacturing footprint
- Establish prioritized action plan for recommendations
The new manufacturing strategy is currently being implemented. Impacts include:
- The operating cost projected savings exceed $100m annually
- The recommendations increase network flexibility and reduce risk
- Our approach is now being used in other business lines across the company
Adapting to the New Normal
In the volatile trading conditions that are emerging, manufacturers that can build balanced, flexible supply and production bases will be best suited to navigate shifting market conditions. But first you need to understand your baseline configuration and its vulnerabilities.
Then you can specify a more resilient production configuration and begin working towards it; progress may be incremental, it need not be a “big bang.” In fact, incremental projects can pay for themselves and establish momentum towards the end state.
But you must begin.
by demo3 | Jan 16, 2026 | Homepage Featured
A recent article (Traditional Risk Models Don’t Apply During Tariff Uncertainty | IndustryWeek) usefully discussed supply chain strategies for volatile trading environments, but it didn’t address some important considerations. The authors catalog short- and medium-term strategies for supply chain management in this much more volatile trading environment, and some of the factors to consider (e.g., product perishability, demand sensitivity).
Yet most of the medium-term strategies are generally not within the purview of supply chain managers (except dual sourcing), no matter how senior:
- Demand management decisions generally belong to sales (even if they consult manufacturing and supply chain)
- Most process agility (e.g., changing fabrication to switch between materials) belongs to other process owners, and the costs to build such capability will be levied against them, not supply chain
- Similarly, manufacturing flexibility is costly for production to implement, and the factories generally own these decisions (and the CAPEX burden)
- Product design lives with engineering, with manufacturing input (we hope)
This highlights a key challenge: effective management of volatile trade environments is both cross-functional and strategic. The medium-term strategies suggested by the authors require functional leaders to coordinate investment and execution if full value is to be realized; it does little good to diversify purchasing to additional materials (such as different grades of steel) if manufacturing isn’t able to accommodate them. The cross-functional nature of these strategies puts a premium on coherent, disciplined decision processes (as the authors point out). Such processes require thoughtful design and testing.
Choices about how to increase supply chain resilience have strategic implications beyond cost, quality, or differentiation — the suggested medium-term strategies can affect key How to Win factors such as order windows, delivery capabilities, product range, production scheduling flexibility, local production, and more. Strategies to improve resilience in the face of volatile trading conditions need to be consistent with the business strategy; and ideally, to enable it.
Finally, companies with manufacturing footprints that span multiple trading regions have a powerful resilience strategy available to them: crafting their production footprint into a manufacturing network. By manufacturing some components in more than one region, companies can improve their resilience to volatility in the costs of labor; materials; shipping; tariffs & trade restrictions; exchange rates; natural disasters; wars; and energy. Such a network can flex to accommodate changing conditions, maintaining production and reducing the impact of costs and disruptions. Fully employing such a network requires some degree of modularity in product design, and the ability to ensure that modules produced at different sites can be successfully assembled in the region of sale; the greater the design modularity, the greater the opportunity.
We designed such a network with a global business unit at John Deere. Engineering was already modularizing product designs (for design efficiencies), and leadership wanted to explore the implications for manufacturing strategy. The resulting production network design showed substantial savings on total production cost (across several demand and cost scenarios), as well as increasing resilience to shocks. Yet, illustrating the cross-functional requirements of such strategies, implementation was slowed by challenges in configuring the ERP to handle the flexing of the network (i.e., shifting production volumes of modules between factories).
For decades, supply chain and manufacturing strategy has been dominated by a single consideration: cost. But in a more volatile global trade environment, manufacturers need to leverage the strategic dimensions of manufacturing strategy to adapt.
by demo3 | Jan 16, 2026 | Homepage Featured
Artificial intelligence is poised to revolutionize manufacturing sooner than many industry leaders anticipate. AI isn’t coming to manufacturing. It’s already here.
While buzzwords like “generative AI” grab headlines, a more consequential force is quietly taking shape on factory floors and in planning meetings: agentic AI — systems that can reason, act autonomously, and drive outcomes based on real-time data and shifting priorities. For manufacturers, this is a once-in-a-generation opportunity to redefine how work gets done across the value chain. This will fundamentally transform how factories operate and how production is actively managed.
Two-Phase Implementation of Agentic AI
The integration of AI in manufacturing will unfold in two distinct phases, with the first already beginning to take shape.
Phase 1: Operations Planning and Scheduling
The most immediate impact of AI is being felt in planning functions — from sales and operations planning (S&OP) to production scheduling. This progression makes practical sense, as most manufacturers already have Enterprise Resource Planning (ERP) or Manufacturing Resource Planning (MRP) systems in place, with Manufacturing Execution Systems (MES) bridging to production systems. These existing platforms provide the necessary data infrastructure and real-time monitoring capabilities that agentic AI requires for optimization.
What’s changing now is the speed and autonomy with which decisions can be made. Autonomous AI agents can dynamically replan based on changes in demand, material availability, or even geopolitical events like trade tariffs or disruptions in shipping lanes. These systems don’t just optimize schedules — they learn and adapt over time, enabling continuous improvement without human intervention.
As an example, Nvidia recently announced at their Global Technical Conference (GTC) a partnership with General Motors to implement Omniverse digital twins for factory planning and, eventually, robotic operations management.
The immediate future will see AI systems updating and optimizing sales and operations plans in real time, responding overnight to changes in orders or sales activities. These systems will transmit signals from the sales end directly to operations scheduling, ensuring that costs, performance requirements, and timing are continuously optimized.
Phase 2: Factory Floor Integration
The second implementation phase—which will take longer due to infrastructure requirements—involves AI integration directly on the factory floor (not just directing operations on the floor). This goes beyond the familiar manufacturing robots in automobile plants to encompass increasingly networked machinery for metal forming, thermoforming, tube bending, and other manufacturing processes.
This machinery is being connected to MES systems that can monitor operations and issue commands based on what they observe. Beyond the attention-grabbing demonstrations of voice-commanded robots, the true value lies in the comprehensive networking of all machinery and operations throughout the factory floor.
NVIDIA’s recent moves are a sign of what’s to come: integrating digital twins, real-time data, and AI to simulate, plan, and manage factory operations. The convergence of these capabilities will unlock massive gains in productivity and agility.
The Ultimate Vision: Autonomous Manufacturing
The end goal is a manufacturing ecosystem where changes in demand or external conditions—such as new tariffs between production and selling regions, spikes in shipping costs, or disruptions to shipping lanes like recent events in the Red Sea—are automatically reflected in sales and operations plans by autonomous AI agents. These adjustments will then cascade to the factory floor operations without human intervention.
While human supervision will remain indispensable, these systems will operate autonomously and continuously, bringing about a revolution in manufacturing productivity that will transform the industry.
This isn’t science fiction. It’s a fast-emerging reality, and the organizations that move early will set the pace for the next era of industrial performance.
by demo3 | Jan 16, 2026 | Homepage Featured
When customers have many choices, companies feel the pressure to expand their product lines. Whether in capital goods or fast food, it’s tempting to add another SKU or slightly tweak a product to meet a customer’s request. This results in a broader range of offerings and increased complexity across engineering, production, sales, and fulfillment. While businesses often try to factor the cost of complexity into their financial projections, this can become its own quagmire. The better approach is to consider strategy directly when managing product line complexity.
The Illusion of Project Economics
At first glance, new products or variants seem like they’ll pay off. However, as the number of offerings grows, the hidden costs of complexity can outweigh any incremental gains in sales. These costs don’t show up in the bill of materials (BOM) but are buried in indirect costs that are allocated across products. The challenge is that these hidden costs are slow to appear and often unevenly distributed. As a result, while individual products may seem profitable, the margins for the entire portfolio can erode over time.
Complexity Can Be Dangerous
Some companies run into serious trouble when their product lines become too complex. For example, Boeing’s decision to offer extensive customization options for the 787 Dreamliner backfired. The resulting production complexity and delays ultimately forced Boeing to suspend production, incur over $9 billion in charges, and lose $45 million per airplane in 2013. In the late 2000s, Dell’s just-in-time model thrived with a narrow product range. But when they expanded to include more consumer PCs, laptops, and peripherals, their operations struggled to keep up. This led to declining profits, stagnating stock prices, and ultimately Michael Dell taking the company private to reset its strategy.
In both cases, these companies strayed from their core product offerings in pursuit of additional revenue, only to be bogged down by accumulating complexity.
The Quagmire of Estimating Complexity Costs
As the burdens of complexity become apparent, operations teams may try to estimate the costs of complexity in an effort to reduce demands for more products and product variants. Yet calculating the cost of complexity is problematic. Many companies hope to simply estimate these costs and incorporate them into their financial models. However, this often becomes a quagmire of complexity itself.
Benchmarking, scenario analysis, and process simulation can help justify reducing complexity, but they don’t always offer convincing guidance on what to cut. Plus, getting product owners to simplify their offerings for the greater good of the company is a tough sell.
Even when cost savings are clear, it’s hard to get teams to commit to complexity reduction, especially if it means limiting their product options for customers. Companies often find that avoiding costs isn’t as motivating as direct cost reductions. And without clear buy-in from product owners, complexity reduction efforts stall.
Climbing Out of the Quagmire: Strategic Scoring
Instead of getting caught up in better accounting for complexity costs in your project financials, live with the cost allocations you have and instead evaluate projects based not only on financial returns but on strategic value as well. A simple scoring model can help. Projects can be rated based on two dimensions: their impact on target market segments (where you choose to compete) and their impact on strategic pillars (how you plan to win in those segments). Products that score highly on both dimensions should be prioritized.
To implement this, companies need a clear strategy that defines where they will compete and how they will win. The “where” is usually determined by prioritized market segments. If you can estimate your market share in each segment, you can put the incremental revenues of a product variant into context: will this variant increase our share? Will it prevent loss of share? The same revenue gains are worth more in a high-priority market segment than in a low-priority segment. Does the proposed product or variant address a high priority market segment? Is it required to defend share in a high priority segment, or to deter a new market entrant? You might accept lower returns (and increased complexity) to defend a key segment against an adjacent competitor, or one moving up the value chain.
The “how” describes how a company plans to beat its competitors in those segments — the strategic pillars. For example:
- Bringing to market an integrated solution suite employing a particular new technology
- Partnering with strategic high-growth OEMs
- Controlling the high-performance end of the market
- Delivering more value than competitors on key product dimensions
Does the proposed product or variant directly help you achieve some of these strategic ends? Is it required? A positive financial return is necessary, but it is not sufficient — not all incremental revenue is worth the trouble.
When you are comparing several projects / products competing for limited funding (or for portfolio simplification), you can score each of them against the others to define relative position on each axis [in the chart below, for a medical equipment maker, each bubble represents a product or potential product]:
Clearly, projects in the upper right are higher priority than those on the lower left, even if they are nominally smaller.
Combining Financial and Strategic Metrics
By integrating financial and strategic metrics, companies can form a more complete picture of a project’s value. For example, we helped a machinery OEM create a one-page scorecard that evaluated projects based on:
- How the new product or upgrade improved performance on the value proposition dimensions of performance, uptime, and cost. For each dimension, we developed objective metrics for comparing current products, proposed products, and competition; projects that delivered the biggest boost over competitors on these three dimensions were favored.
- Which priority market segments would be impacted, and how
- Financial metrics, including volumes, sales, ROI, and R&D costs
This scorecard helped the company develop the products that best fit the value proposition and avoid me-too projects that didn’t align. Complexity in the product portfolio (such as product variants) was materially reduced over time — by focusing on strategic value, in addition to financial metrics.
Over the development cycle after this value proposition (and enabling scorecard) were introduced, the OEM’s revenues increased by over 50% while net income more than doubled — despite an intervening industry recession.
Don’t Get Distracted
Focusing too much on calculating the costs of complexity can be a distraction. While it’s tempting to distill these costs into a single number for financial models, this oversimplification can lead to bad decisions. Companies that align their product decisions with both financial expectations and strategic goals tend to outperform over the long run. The winning approach is to avoid getting bogged down in cost debates and stay focused on the strategic factors that matter most.
