Supply chain leaders are under frequent pressure to contribute to cash flow by reducing inventory. This research provides a set of sequenced actions for a comprehensive approach that will increase cash flow while protecting service levels, profitability and growth.
Companies that struggle with inventory reduction often lack understanding about the impact of decisions on specific inventory functions. They also overlook the need to balance working capital efficiency with incentives for reducing unit supply cost.
Companies can become obsessed with modeling to determine precise inventory entitlement targets rather than taking the first important steps to reduce inventory.
Companies try to make multiple simultaneous changes in pursuit of inventory reduction, making it difficult to measure the impact and manage the risk of each individual change.
Supply chain leaders under pressure to reduce supply chain inventory should:
Create a sequenced plan for inventory reduction that accelerates low-risk changes and preserves the ability to measure the impact of each change.
Enable the transition to pull-based supply by focusing first on changes to neutralize the forces that drive large order quantities and oversupply motivated by low-cost mindsets and incentives.
Address structural network changes and realignment before implementing advanced inventory buffering and demand management analytics.
Every business has unique product characteristics, supply network attributes and market dynamics that impact its need for inventory. As a result, a roadmap or playbook to reduce supply chain inventory will be unique to each business. The combination and sequence of changes to reduce inventory must take constraints, risks, complexity and value into account.
The probability of achieving and sustaining success will be enhanced by developing and applying inventory intelligence to the decisions that impact inventory, which are summarized within the framework of Table 1. (See )
Achieving and sustaining success requires that the insights from Table 1 be converted into changes in the way decisions are made based on knowledge and judgment about their impact on inventory. These range from design choices about the business itself, including the extended supply network, cost and value trade-offs in the operation of the network, and situational business decisions that balance opportunity with risk.
Supply chain planning and strategy leaders with performance imperatives to reduce inventory can use this list of recommended actions as a checklist to ensure a comprehensive approach that combines conventional changes with more advanced techniques. These should be evaluated and translated, based on the business environment, into an appropriate sequence of changes that reduce inventory while achieving target outcomes.
Sutton's law has relevance to the reduction of cost and working capital, even though its origins are mythical. The infamous criminal Willie Sutton denied saying that he robbed banks because "That's where the money is."
We can use Figure 1 to apply Sutton's law by prioritizing actions for inventory reduction where the balance between opportunity and risk is most favorable. This framework illustrates how supply objectives can vary based on three dimensions of demand within the product portfolio of a business: volume, variability and value. As a result, supply and inventory planning tactics must be differentiated to achieve desired results at acceptable risk. For example, optimization can only be applied confidently in Segments 1 and 2, where demand is reasonably stable and predictable (as measured either by the coefficient of variance [COV] or forecast root mean squared error [RMSE]). Although demand and inventory volume might not be high in Segment 4, management of obsolescence risk increases in importance, offsetting the benefits of minimizing total unit supply cost. Event-based peak demand in Segment 3 requires scenario analysis and business judgment to balance opportunity with risk.
Sutton's law would have us prioritize actions in Segments 2 and 3 of Figure 1 (as opposed to Segment 4). This allows higher prioritization of actions, which will take pressure off the generation of additional excess inventory, which consumes cash and further increases obsolescence risk. Although items in Segment 1 may have lower impact on total inventory, these might represent quick wins with low risk. Automation of replenishment decisions for these items can also bring value by increasing planner productivity to enable increased focus in other segments, which involve greater risk and opportunity for impact.
Table 2 contains a list of actions that reflect the above considerations for the pursuit of reduced inventory. These actions are intended to address all forms of inventory (e.g., raw material, finished goods and work in process), while protecting service levels, product margins, operating efficiency and growth opportunities.
Neutralizing the root causes of oversupply reduces the cost and risk of aging inventory, with a potential balance sheet boost for the stock price. This step is essential to enabling the implementation of pull-based replenishment to finished goods warehouses in the channel, which can then be automated with analytics for visibility and digitalization.
Challenges and Trade-Offs
Reversing the forces of oversupply requires substantial influence and persuasion skill to neutralize existing mindsets and organizational incentives. A strong supply planning role (with accountability for inventory control) is essential to anticipate and avoid decisions that will result in excess inventory that becomes difficult to eliminate once it has been established.
No. 2 provides an opportunity to push past the high-level hand-waving associated with strategic inventory and drive for clear decision governance and supporting rationale for situational inventory needs. This also creates a clear and beneficial distinction between operational decisions that can be analyzed and proposed by supply chain versus those that defer to business judgment but also call for accountability, documentation and review.
No. 3 Consolidate Finished Goods Stocking Locations
Reducing the number of stocking locations can have a beneficial impact on total inventory levels. One company reduced its warehouses by 34% for a 23% reduction in total inventory. This is a structural change to the supply network that requires careful modeling and organizational alignment to determine the right balance between design trade-offs for service, cost and resilience. Consolidation can be selective rather than uniform across the product portfolio, based on segmentation analysis (see ).
Aggregation of demand into fewer stocking locations has a beneficial impact on warehousing cost and forecast accuracy for lower safety stock requirements. Cycle stocks may also be reduced for low-volume SKUs challenged with minimum order or replenishment quantities. Fewer stocking locations simplifies network operations and demand fulfillment decision making. Segments 1 and 4 of Figure 1 would contain items more likely to benefit from pooling and warehouse consolidation.
Challenges and Trade-Offs
Consolidation of stocking locations must be tempered against minimum order lead time requirements.There could also be a freight cost trade-off to fulfill certain demand across longer distances.A more complicated segmented approach may be required, in which low-volume items are consolidated into fewer locations while larger volume items are stocked at more locations. This can complicate order fulfillment for items frequently ordered together.
No. 4 Optimize Pull-Based Replenishment to Finished Goods Warehouses
Pull-based replenishment of warehouses represents the most basic opportunity for postponed supply.This can be implemented once stocking locations have been consolidated and the forces behind push-based replenishment have been neutralized.Reorder points can be determined based on lead times, demand patterns and service-level targets using established safety buffer analysis (see ). This allows product to be stored centrally (at factories and central warehouses) before it is committed to a specific region or local distribution center (DC).
There is no need to take undue service risks during initial implementation. Safety stocks can be sized liberally to match service-level targets and avoid premature internal conflict. The focus should be on using these generous safety stock allocations to replace human judgment and control warehouse replenishments based on analytics to maintain target service levels.
A conversion to pull-based warehouse inventory replenishment makes it possible to measure performance within a target range using the inventory health metric described in The reference ranges for each location will be based on some tolerance allowance from the safety stock level (minimum target) replenishment quantity plus safety stock (maximum).Processes and governance for exception management can follow the implementation of this analysis, including monitoring and mitigation of aging inventory situations in collaboration with commercial roles.Many companies are lacking defined governance and decision guidelines for obsolete inventory. Supply chain's primary role is to provide the visibility with analytics, facilitate the review and document recommendations for action.
Excess and aging inventory in the warehouses will be reduced with a pull-based approach. Replenishment planners can be held accountable for maintaining inventory levels within targets. This step is a precursor to the implementation of more advanced buffering analytics (see ).
Challenges and Trade-Offs
Safety stock inventory represents a balance between service and cost.Avoid the fear of service failures by initially implementing generous service-level targets.Focus on control in the target range and observe performance.Conduct analysis to identify items with opportunity for better performance based on better demand management.Consider consolidation and centralization of the replenishment planning function, with a focus on master data management, performance monitoring, exception management and improvement across the entire distribution system.
No. 5 Implement MRP-Based Flow of Inbound Parts and Raw Materials
Manufacturers and service providers must develop capabilities for planning the supply and inventory of raw materials and parts that support their operations. Raw material and parts inventory is part of the overall supply strategy (rather than managed and evaluated in isolation). Figure 1 can be translated and applied to differentiate how inventory principles such as optimal order quantities, postponed commitments and risk mitigation with hedging stock are applied to raw materials and parts. Segmentation criteria should consider more than spend volume
There is a unique relationship between item value, supply lead time and other production operating strategies for each direct material in the product bill of material (BOM). This creates the need for customized architecting of a material supply and inventory management plan for each material or part. This will include priorities for supplier integration, collaboration and lead time optimization where improvement opportunities are greatest.
Applying inventory knowledge in a differentiated way based on segmentation analysis ensures that raw material and parts inventory supports finished goods manufacturing and service delivery in a way that maximizes value and manages risk.
Challenges and Trade-Offs
Material planning has been embedded within manufacturing operations traditionally, constrained by purchasing tactics and expected to ensure efficient asset operations without applying a broader view of inventory risk or product/service value. Material planning in many organizations requires the application of new thinking, technology and talent to maximize its value. There is no single solution that applies to every operating environment, requiring careful synchronization with the product and supply chain. Long supply lead times and complex BOMs add significant complexity to material planning.
No. 6 Reduce Structural Inventory by Mitigating Critical Network Constraints
Every network has unique and complex combinations of constraints that impede flow and occupy inventory. More importantly, these constraints impair the agility of the supply response to demand signals and opportunities. The benefit of eliminating constraints is better agility and service performance. The reduction of structural inventory associated with these constraints is more of a dividend than a driver for the change, in most cases. Examples of constraint elimination include:
Extra capacity and improved scheduling for postproduction quality testing, compliance processes and outbound loading docks can shave multiple days of finished goods inventory that is waiting to be released for sale.
Expansion of assembly and packaging lines (or other shared capacity), which often becomes a bottleneck, can take pressure off waiting and release substantial semifinished goods inventory from work-in-process status.
Factory floor layout adjustments that create streamlined movement of jobs through a flexible shop floor can reduce work in process by faster travel time through the factory.
Use of expedited freight modes (under certain conditions related to product value and obsolescence risk) can yield a favorable trade-off between cost and working capital.
Constrained production scheduling and synchronized production planning across sites involving interdependent multistep production reduces queues of material waiting in between process steps. (See )
The main driver for improved network agility is improved service performance to support customer experience and growth. A reduction in structural inventory and obsolescence risk is a nice dividend but rarely is sufficient as the main driver.
Challenges and Trade-Offs
Increased capacity, shorter lead times or added flexibility involve a cost that must be weighed against the network performance benefits, including lower structural working capital.This requires organizations to rethink legacy design principles that always prioritize the lowest total cost of supply, regardless of its impact on network capability.
No. 7 Align Supply Planning With Inventory Targets
Lower inventory targets do not translate into reduced inventory until one or more of the six preceding actions have been taken. Inventory targets in the supply plan can now be lowered, incrementally from current levels, to bleed stock out of the network. There will still be a number of considerations to balance in terms of the rate and extent of inventory reduction, including the impact on capacity utilization and supplier contracts (see ).
The degree of inventory reduction that is possible (to new and lower operating levels) will depend upon the constraints that have been removed by actions described previously. The new operating levels can be estimated analytically, involving data intensity and modeling complexity that corresponds to the details of the network, product portfolio and demand behavior. These entitlement estimates will involve simplifying assumptions that still require validation during implementation. The alternative is to use empirical data associated with service levels and operating efficiency that tunes inventory target levels based on measured performance of the network.
Inventory reduction enabled by preceding actions is realized and sustained through ongoing supply planning and closed-loop performance management.
Challenges and Trade-Offs
A single supply plan must be established and managed for each network segment, potentially requiring role, process and system model changes. This includes increased measurement and pursuit of supply plan performance for all factories, warehouses and suppliers.
No. 8 Consolidate Product Designs, Obsolete Products and Common Items
Item proliferation is present in all industries from discrete manufacturing to healthcare. As portfolios become more expansive for more specific targeting of customer and market segments, the ability to predict SKU-specific demand (including location and timing) deteriorates. Inventory can be used to buffer some variability and uncertainty. However, the risk of excess and obsolete stock grows in the long tail of niche or customized items that have not established demand across a sufficiently broad portion of the market.
There is often duplication in common purchased service and repair items across multiple locations or regions of an organization. If technical requirements and product specifications can be standardized, then these can be consolidated into a single item to create both purchasing leverage and inventory consolidation. Once the obvious duplicates have been eliminated, a more comprehensive process must be established. Consolidation of finished products involves more stakeholders and complex considerations, such as customer-specific products and specifications. Leading supply chain organizations are playing a key role in cross-functional efforts by bringing data and analysis to bear to support growth at sustainable costs (see ). Key success factors to capitalize on this cost reduction opportunity include:
Designate a team to begin looking for product portfolio improvement opportunities. Establish realistic unit cost expectations for low-volume items in the early and late stages of product life.
Look for negative KPI trends, such as supply costs accelerating faster than revenue, to justify the need for further investigation into the causes. Also, link portfolio optimization to positive business outcomes such as profitable growth, faster inventory turns and improved service levels.
Attribute activity-based supply chain costs to individual products to make better-informed decisions on when to launch, refresh or remove products from the current portfolio.
Compare inventory write-offs and margins at a SKU-level to identify products where write-offs exceed or significantly minimize margin contribution (see ).
The main benefit of these consolidations will be the creation of supply leverage and sourcing efficiencies for purchased material and finished products alike. Reduced inventory obsolescence risk will be the primary benefit for finished goods (part of an overall focus on cost reduction and profitable growth).
Challenges and Trade-Offs
There are often technical and regulatory constraints to making these consolidations.Commercial considerations include the existence of complementary products that are part of a broader offering and should be valued based on their contribution to an overall portfolio.Supply chain's largest barrier is in achieving the status of business partner that can successfully bring these opportunities for consideration and action.
No. 9 Design and Implement Postponed Processing
Postponement is an easy concept to embrace, but requires a comprehensive approach from design through scheduling and execution. Postponed processing shifts inventory exposure from finished goods to raw material and semifinished intermediates to support late-stage differentiation of product configuration. Total inventory value may go down, but reduced obsolescence risk (while maintaining competitive order lead times) is the more likely driver for this change. However, finished goods buffer is replaced with the need for a capacity buffer and rapid scheduling and execution capability to convert sales orders to completed production orders. Postponement cannot be thought of as a lower-cost form of supply, but one that supports finished product variety and competitive service levels while minimizing inventory obsolescence.
High-tech manufacturers leverage the fact that forecasting is more accurate at the product family level than the customized order level. They manufacture the "base" unit and build the final configuration when forecast accuracy improves closer to the time of the order placement. One leading producer of flash memory had to work with its design teams to make product modifications that enabled postponement. Storage capacity for raw material and intermediate products must also be designed into the network.
Supply chain leaders should map the points in time where the product goes through differentiation activities as it moves through the value chain. Identify opportunities to delay differentiation when the delay leads to improved information about demand as measured by changes in forecast accuracy. Consider products with long supply or manufacturing lead times as candidates to evaluate in order to gain benefits related to the risk of producing against a biased demand plan (see).
The benefit is the optimization of total inventory cost and risk in support of configured products and competitive order lead times.
Challenges and Trade-Offs
This action requires inventory in a different form (raw materials, intermediates and packaging materials) and iscontingent on design changes to the product and network to enable postponement and intermediate storage. It requires excess capacity and agile scheduling for the last stages of assembly in response to orders.Operating a mixed model that blends MTS and postponement involves more complex scheduling requiring advanced capabilities.
No. 10 Implement Advanced Buffering Analytics
More mature organizations with complex distribution networks or constrained manufacturing processes that have worked through many of the items above should consider the development of advanced stock buffering analytics. Finished goods distribution is often a good place to start the multiechelon inventory optimization (MEIO) journey (see ). Take a pilot approach that includes products with different demand characteristics (e.g., predictable, slow-movers with higher variability). Understand where the value of changes to current policies resides because the value will often be higher for higher-volatility products.
MEIO buffering analytics also support multistage production processes, including postponement strategies, to determine the location and extent of buffering to protect capacity constraints and support service levels (see ). Each implementation of MEIO is a custom implementation requiring high levels of process integration, data quality and adoption change management.
Benefits include reduced inventory by avoiding redundant buffers, better management of complex mix for improved service, and the ability to optimize inventory mix and capacity utilization within an overall (average) service level or total inventory target.
Challenges and Trade-Offs
This action requires investment and attention to data quality and integration. Recommendations are counter-intuitive and require change management, including automation to leverage the benefits and protect them from human bias and override.Failure rates are high if there is not organization readiness for adoption and change.
No. 11 Develop Advanced Demand Sensing and Modeling Analytics
Another advanced approach is the development and implementation of advanced demand sensing and modeling analytics. Demand modeling can deliver improved forecasts that mitigate the impacts of forecast error, including lower inventory safety stock.
Demand sensing captures short-term demand signals or predictors (e.g., POS, weather and IoT) that reduce demand latency for more efficient and effective operations scheduling. This improves the overall synchronization of operating processes to reduce lead times for more responsive service while shaving away portions of structural inventory and operating waste.
Not all products require or benefit from demand sensing. The capability must be matched with agile supply that can capitalize on more frequent updates to the demand signal. Campbell invested in near-term demand sensing technologies and used SKU demand profiles to determine which supply plans would benefit from daily statistical forecasts (see ).
Benefits for this are better service and inventory levels at the operational level, and better situational decisions and reduced overall safety stock based on demand predictive modeling.
Challenges and Trade-Offs
This involves investment with uncertainty of results and benefits. Analysis is required to develop targeted applications where value of these analytics is highest.Supply agility is required to take advantage of demand sensing visibility in the near-term horizon. Insufficient supply chain analytics talent can be a barrier to developing advanced demand models.