This blog post covers business challenges and recommended supply chain excellence strategies to deal with them for Automotive OEM companies, specifically focused on the Automotive Spare Parts Supply Chain.
When we put together the increasing strategic importance of spare part business with the inherent complexity in managing this business, we can appreciate the complex challenges and the large business value of spares supply chain excellence for Auto OEMs. The best in class performers will manage their spare parts business in a forward looking and advanced analytics driven manner when it comes to spare parts demand forecasting, inventory optimization, and integrated business planning across the multi-enterprise landscape.
I will cover the following areas in this post for Automotive Spare Parts Businesses:
1. Growing Strategic Importance of Spares
2. Maturity Stages: Crawl-Walk-Run
3. Organization Structure Choices
4. Multi-tier Distribution Network
5. Forecasting & Planning Challenges
6. Forecasting & Planning Strategies
7. Supply Chain Performance / KPIs
8. Closing Remarks
Growing Strategic Importance of Spares
Superior after-sales service is now widely accepted as a key driver of new vehicle sales in a consumer driven world of exploding choices and diminishing brand loyalties. Overall automobile market size continues to expand in developing countries albeit with slowing growth rates. As the average vehicle's age in the market increases, spare parts and service business is increasing its share of total revenue at Automotive OEMs. Spare parts business is emerging as a key focus area for Auto OEMs given its direct influence through spare parts growing revenue with high profits and indirect influence on vehicles revenue as a loyalty enhancer. It's a win-win for Auto OEM executives and End Customers as spare parts operations excellence directly translates to both higher overall profitability as well as higher consumer satisfaction with service quality and resulting boost to branding. This is significant as domestic OEMs compete with international brands for slowing growth based overall market size projection. Service excellence is key to customer retention. Supply chain excellence is key to doing so profitably.
As developing markets mature in terms of vehicle sales cooling down, the spare parts business will increase its strategic importance given the typical high profitability associated (76% higher than vehicle profitability) along with the higher revenue share over the years. In mature markets like US and Europe, spare parts business is increasingly the main driver of profits as it brings in 25-40% profit margins and accounts for 35-50% or more of OEM revenues. In contrast, OEMs in developing markets only produce 5-10% revenues from their spare parts business. However, the profit margins are 20-30% and spare parts business growth rates are in excess of 25% making spare parts business a strategic focus area. We encourage domestic OEMs in large developing markets such as BRIC countries to strategically reposition the spare parts business and pay special attention to balancing cost, service levels, and profitability from spare parts operations.
Maturity Stages: Crawl-Walk-Run
Similar to supply chain planning maturity stages from leading market analysts like Gartner, spare parts operations have their own 3-level crawl-walk-run maturity cycle.
- Stage 1 (Crawl): Spare parts business is operated by functional departments within the vehicle centric organization structure. It is difficult to get any economies of scale, service levels are volatile, spare parts business is a cost center and leverages the Vehicle Supply Chain Management, Planning, and Procurement organizations. Managers are focused on achieving service levels (reliable supply availability) at any cost. Cost and profits are of secondary importance, focus in on surviving the brand and providing a floor service level. Sourcing and supplier management is managed by the Production department. Combined Procurement and Inventory Management across parts inventory for production and aftermarket spares purposes leads to random service level issues. Spare parts planning is not aftermarket demand specific and hence cannot react to rapidly changing market demand.
- Stage 2 (Walk): A network effect becomes evident as spare parts business grows rapidly, distribution network matures, and service levels become predictable. Spare parts business is as a cost center, although revenue & profits are now being tracked and the spare parts contribution to overall business is becoming noteworthy. Focus moves to supply chain synergy and flexibility across vehicle production oriented parts demands and spare parts demands. This stage is still focused on service level, customer experience, and growth. The focus has not moved to profitable growth. Customer managers in this stage intuitively realize the high costs of service, but are resigned to it as cost of doing business.
- Stage 3 (Run): Spare parts business formally gets recognized as a profit center and becomes its own business unit to deal with spare part sales and delivery, given the significant revenue and profits this business is bringing in. Cost and profits become center stage now given the service level has stabilized. Better demand prediction, optimal inventory levels, supply chain agility to handle shocks, supplier and distributor collaboration, multi-enterprise supply chain visibility all become enablers of profitable growth as managers focus on profitable growth with guaranteed service levels.
Stage 1 and 2 OEM environments are characterized by high neglect of the spare parts business profitability, required business positioning, and internal stakeholder alignment around how to effectively manage spare parts business. We recommend auto OEMs to look at their organization structure for spare parts business and align it strategically with the business strategy as they look to drive brand loyalty and enterprise profitability which can be achieved through a well-managed and profitably growing spare parts business. Most domestic OEMs in emerging markets belong to stage 1 or 2 as they adopt a push sales model in domestic markets aimed at increasing revenues. These environments are known for their inaccurate forecasting & inventory planning and limited supply chain visibility. These environments achieve somewhat respectable service levels through inordinately high inventory levels, unresponsive order to delivery lead times that hide planning and operations problems in the spare parts supply chain, but increase the risk of customer satisfaction and experience being sub-par leading to eroding brand loyalty.
For stage 3 OEMs, that have established their Spare Parts Business as an independent subsidiary, most do not have advanced analytics support to optimally maintain the balance between cost, service level, and profitability.
Organization Structure Choices
There are multiple organization models available for Automotive OEMs to set-up their spare parts business:
- Functional: This is typically associated with stage 1 OEMs. In this model, the organization is set-up functionally in terms of Centralized Supply Chain Planning, Logistics, Sales, Production across vehicles and spares. This leads to poor agility of spare parts business and high cost of inter-functional coordination.
- Hybrid: This is typically associated with stage 2 OEMs. The spare parts sourcing & procurement function is left with the Vehicle Production company as the spare parts sales, planning, and distribution logistics are integrated within the Vehicle Sales company.
- Independent Subsidiary/BU: This is typically associated with stage 3 OEMs. As the spare parts business is set-up as an independent BU/subsidiary, the subsidiary has its own sourcing and planning processes independent from the production parts demand. When the Vehicle Sales and Vehicle Production are set-up as two separate companies, the spare BU may also be created within the Vehicle Sales company. This increases focus and agility, but struggles to maintain economies of scale that a stage 2 offers (production demand and spare parts demand pooled together) and also struggles to balance cost, service, and profitability. Volkswagen and Ford have independent BUs to manage spare parts operations. Hyundai and Chrysler have their own independent subsidiaries (Mobis and Mopar respectively).
Multi-tier Distribution Network
Spare parts supply chains have high diversity on both ends: in terms of suppliers and type of spare parts on one hand and the high volume of distributors / dealers / service stations on the other hand. Distributors supply parts to the service stations (branded as well as non-branded) and dealers buy direct from the OEM. There is also dealer to dealer and distributor to dealer sales that happen to chase high service levels or as players offload their excess inventory to other players in the ecosystem. Branded and non-branded service stations may also be motivated to buy grey market parts, parts which are identical in terms of function, but not from the OEM or authorized part suppliers. From the OEM side, there is a Central Distribution Center (CDC) which services Regional Distribution Centers (RDCs). Some small-scale OEMs may have just a one-tier distribution network.
There is another dimension to the distribution network design: buyout vs. self-run models.
- Buyout models are typically deployed by domestic OEMs or regional players in international markets, where they rely on large scale distributors to sell and distribute spare parts in their respective markets. Distributors buy spare parts from the OEMs, then distribute to dealer in their market. Distributor inventory or distributor sales patterns are not typically visible to OEMs in the buyout model. This leads to increased service level and hence, customer satisfaction and overall brand risk as distributors make inventory stocking decisions based on their profitability as they stock fast moving parts while sacrificing service level of medium/slow moving parts. OEMs need less working capital, realize revenue faster, and need for spare parts optimization is less intense than self-run model. The additional echelon in the supply chain (wholesaler / distributor) degrades supply chain information, visibility, and synergy. The bull-whip effect gets more intense as individual players make disconnected plans from their perspective. This results in higher inventory levels across supply chain echelons and reduced inventory efficiency.
- Self-run models are typically deployed by joint-venture OEMs as they take full responsibility of providing spare parts into a given market. This is done by establishing Regional Distribution Centers (RDCs) or by working with Third Party Logistics 3PL providers to support regional distribution. OEMs own all the spare parts inventory in RDCs. Self-run model provides full visibility into dealer demand patterns (how dealers buy from RDCs), Visibility into true dealer demand enables accurate demand sensing and inventory optimization through the echelons (Dealers, RDCs, CDC, and Suppliers). Service level can be tightly maintained across all classes of spares (not just fast moving) and enables overall optimization of the spare parts supply chain. However, it increases working capital requirements, overall inventory and supply chain cost / complexity to the OEMs.
Self-run models are better in the long run as they reduce an echelon (distributor) from the supply chain and brings it back to Supplier-OEM-Dealer supply chain. It enables accurate demand & inventory planning, increases visibility, and enables lean and optimal supply chains that best balance cost, service level, and profitability. For OEMs in the buyout model based distribution networks, the transition to self-run needs to be managed carefully given the risk to business as distributors sell both vehicles and spares. Self-run models also require a robust level of forecasting & planning capability to be able to take advantage of the proximity to market and true demand signal visibility that this model affords.
Forecasting & Planning Challenges
Many Auto OEMs are still ways away from being able to provide cost effective and excellent service & spare parts availability experience to their end customers. Given the increasing expectations of flawless service that consumers implicitly bring to service shops, all it takes is a couple bad experiences for the end-customer to switch brands next time they are in the market for a new automobile. What's at stake is customer lifetime revenue, most OEMs realize that, but feel hampered by inherent complexities in providing desired service levels in a cost sustainable manner. As many OEMs are now providing 10 years / 100,000 miles based standard warranties to lure customers, spare parts operations need to be optimized to ensure warranty costs remain under control to manage overall profit margins for the enterprise.
Typical complexities include erratic parts supply of spares for old models, lack of supply chain visibility across multiple echelons, difficulty in assessing optimal inventory levels with available inventory planning capability, complicated End of Life EOL scenarios with complex supersession and replacement rules, trading partner collaboration challenges with both spare part suppliers, dealers, and distributors in a buyout model, operational planning and execution oriented information systems with basic forecasting & inventory planning capability, organization culture that believes in pushing dealers to buy spares to meet revenue targets, or rewards fire-fighting & expediting over careful planning.
Let me provide more color to the challenges as this is a complex problem we are dealing with. SKU portfolio is segmented into serviceable and non-serviceable parts. Non-serviceable parts are no longer procured and existing inventory is depleted through dealer orders. Another segmentation is current vs. non-current parts. Non-current spare parts represent spare parts demand for which there is no corresponding vehicle manufacturing oriented parts demand. This happens when a particular model is no longer manufactured, but needs to be serviced for another 7-10 years. In some cases, individual spare parts need to be ordered from different suppliers (kit child parts), then they need to be put together in a kit to form a saleable spare part. Forecasting needs to be at kit level with Bill of Material BOM explosion to procure individual kit parts. Non-current and serviceable parts need a minimum order quantity for the supplier to run profitable manufacturing operations given the lack of vehicle manufacturing oriented parts demand in this scenario.
Supply chain planning for spares supply chains is considerably more difficult as compared to vehicle supply chains given intermittent demand patterns and exploding spare part Stock Keeping Unit (SKU) portfolios. Intermittent demand is typically skewed and laced with high demand variability. This results into high supply chain inventory and noise with traditional demand forecasting and inventory management systems available from generalist ERP software providers. The safety stock calculation logic based on normal distribution of variability is not effective in case of skewed demand.
Spare parts life cycle is much longer than vehicles life cycle given the mandate to serve the market with spare parts for an additional 7-10 years after the corresponding vehicle model has been retired. Spare part SKU portfolios are also exploding given international market specific variants and proliferation in vehicle SKUs. Spare parts have high service level requirements for a broad Spare SKU portfolio of 25,000 SKUs or more. Pareto's principle is in full force here as top 80-95% of revenue comes from top 4000-6000 SKUs. This means a very long tail, which still needs to be provisioned from an inventory perspective to provide a decent service level without excessive inventory investment and write-off risk. The inherently unpredictable nature of spares part demand combined with this long tail presents a significant inventory problem. Traditional Service Parts Planning solutions do not address this long tail of parts effectively to provide optimal service levels across channels.
Many Auto OEMs find the following as major barriers to service excellence: Planning & Forecasting, Multi-echelon Inventory Management, Supply Chain Visibility, inadequate / inflexible IT systems. This is the low-hanging business opportunity we are talking about given some of the other factors like supplier delivery performance / reliability, long lead time for purchased components, supplier relationships, etc. are much harder to resolve.
Forecast accuracy for spares at SKU/RDC level is 50% or less in most cases. Forecast accuracy at lead time further deteriorates for regional OEMs exporting to international markets given lack of true demand visibility in buyout models and long lead times to get spare parts shipped through sea and to clear customs.
Many Auto OEMs still think of demand forecasting as a past history extrapolation exercise into the future. They use last 2-3 years of shipment history and simplistic math (weighted moving average or exponential smoothing models). Some OEMs rely on manual forecasting approaches asking their dealers, sales teams, or supply chain planners to forecast future spares demand. The manual approaches cannot scale the full Spares SKU portfolio (tens of thousands of SKUs), cannot reduce demand latency as they are typically monthly cycles, and are fraught with human bias risk as sales teams bake in their targets and push sales plans as a ‘forecast’ of future demand.
These environments lack a forward looking causal demand modeling based systematic and scientific approach to sensing demand across the full portfolio with minimal demand latency. Instead, we usually see a monthly forecasting cycle supported by simplistic demand forecasting techniques from the ERP vendor, run at monthly frequency which introduces unnecessary latency and error that needs to be covered by higher levels of inventory or order to delivery lead times which lead to unresponsive or expensive supply chains.
Many Auto OEMs resort to a push model for spares as they believe dealers will resort to grey market purchases if they experience stock-outs even when order to delivery service lead time is not a challenge. Dealers' erratic ordering patterns further complicate the modeling and generation of an accurate demand forecast of future buys from dealers. High supply chain agility is required given the lack of maturity in forecasting & planning function.
Many Auto OEMs have a Dealer Management System, which enables them to see dealer level spare parts true demands through job cards, dealer inventory, and dealer demands on RDCs. While this level of visibility is great, the intermittent nature of spares demand combined with the options dealer has to service it (buy from OEM, buy from distributor, buy from another dealer, buy grey market parts) make it difficult to accurately forecast dealer demands on to the OEM. Most OEMs are in the basic maturity stages when it comes to demand forecasting, inventory optimization, and integrated business planning to balance cost, service level, and profitability.
Many Auto OEMs lack sophisticated inventory optimization tools and resort to human experience based "one size fits all" approach to establishing days of inventory cover across entire classes of Spare SKUs (fast moving vs. slow moving).
Most OEMs have completed their ERP implementation and warehouse management implementation. Most OEMs are limited by a basic level of capability when it comes to demand forecasting, multi-echelon inventory optimization, and integrated business planning to optimize multi-enterprise or OEM enterprise as a whole based on a set of operational, commercial, and financial constraints, and chasing a specific financial objective like net margin constrained by a floor specific service level.
Deployment planning is complex given horizontal movements of stock to satisfy demands (dealer to dealer, distributor to dealer, grey market supplies, etc.). OEMs that combine production and spare parts demands for procured parts struggle in face of huge dealer vehicle forecast variations month-on-month which can result into the common inventory pool being poached by vehicle production at the expense of spare parts service levels. Suppliers insist on minimum order quantities for older spare parts, especially ones for which there is no vehicle manufacturing demand (recently retired model, market still needs spares). This creates imbalance in demand-supply leading to excess inventory. Spare parts can also enter through remanufacturing operations leading to complexity in planning inbound supplies.
Given the low maturity forecasting and inventory planning capability available to most OEMs today, planners have to resort to manually overwriting release schedules to suppliers resulting in uneven plan quality depending on the planner as well as unproductive business operations in planning & procurement with frequent changes and back-and-forth with suppliers.
Very few Auto OEMs are able to perform advanced network modeling for optimal design, transportation forecasting & optimization, or an end-to-end robust what-if simulation to evaluate multiple trade-offs across commercial, operational, and financial domains.
Forecasting & Planning Strategies
To solve this spare parts planning problem effectively, we need accurate demand modeling / sensing capability to forecast intermittent demand accurately at a granular level combined with a multi-echelon inventory optimization capability. Instead of extrapolating monthly bucketed shipment history into the future, GitaCloud Demand Sensing software helps analyze demand history at SKU / RDC / Customer order line level across the entire Spares SKU portfolio in an automated manner to generate the signal on a daily basis. We help model the life cycle impacts (New Product Introduction NPI / End of Life EOL / Part Supersession & Replacement scenarios).
Automated demand modeling is required to understand vehicle populations and age by model years, understand failure rates through regression techniques across the entire Spares SKU portfolio, with a full understanding of regional diversity. For example, parts fail differently in coastal humid areas where risk of corrosion is high. Some OEMs have started to do this level of forward looking causal modeling for top 500 or 1000 SKUs, but this is nowhere close to their full SKU portfolio that needs to be forecasted and the manual approach cannot scale to be more frequent than a monthly frequency. Another demand influencer can be the type of engine, as parts related to a 100 cc engine may fail differently from parts related to a 150 cc engine.
GitaCloud Demand Sensing software leverages Dealer Point Of Sale POS data through job cards, channel shipments data to dealers / distributors at order line level, and other demand influencers like engine type, or region with high humidity to crunch data in a highly automated manner to produce a highly accurate demand forecast signal using industry leading predictive analytics capabilities. The resulting signal is order of magnitude better (40% or more reduction in forecast error) than traditional monthly history extrapolation or manual forecasting by dealers / sales team approaches.
OEMs need to clearly think though tactical planning, operational planning, and operational execution. Demand Forecasting and Inventory Planning recommendations come from tactical planning. Demand sensing, inventory optimization - rebalancing across RDCs, dispatch, and procurement plans come from operational planning. Procurement and Stock transfer execution belongs to the operational execution. OEMs need to bring in advanced analytics (predictive analytics for demand sensing, prescriptive analytics for inventory levels and optimal balance of cost, service level, and profit).
Inventory requirements should be calculated in an SKU specific manner based on demand volatility, lead times (domestic vs. exports), and desired service level for a specific SKU. Inventory optimization should also be undertaken in a multi-enterprise manner with optimization of dealer inventories, and factoring in supplier capacities and minimum order quantities. Auto OEMs should position inventory as a business strategy lever to offer program specific service levels with service level optimization instead of the traditional ABC classification.
GitaCloud Multi-Echelon Inventory Optimization software operates on top of the accurate demand signal to optimize service levels by SKU, replenish efficiently as it reduces inventory levels and the bullwhip effect throughout the network, while improving planner productivity.
GitaCloud Integrated Business Planning software can provide an optimal solve at the enterprise level while respecting a wide range of operational, commercial, and financial constraints. Companies can choose from a wide range of objective functions: market share, revenue, profit, return on assets, and so on. The optimal solve takes the strategic objectives of the organization; policies associated with the end-to-end supply chain across inventory, service level, logistics, etc.; financial costs (activity based costs, non-linear volume-price curves, etc.).
Supply Chain Performance / KPIs
What does best in class performance look like when it comes to managing spare parts supply chains in Automotive sector? Best in class OEMs are able to get 98% fill rates (% of order lines filled by customer facing warehouse), 6-8 inventory turns, 1 day or less order-to-delivery lead time for greater than 95% of orders, and logistics costs (outbound transportation costs and warehousing costs only) at 4-5 % of sales. Compare this against average performance of 93% or lower fill rates, 3-4 inventory turns, 3-5 days order-to-delivery lead time, and logistics costs at 7-8% of total sales to appreciate the vast gulf between average and best in class performance.
We encourage Auto OEMs to look hard at their spare parts KPIs to see where you are situated and how advanced analytics based forecasting & planning can take you forward. ROI on spare parts optimization is significant given how excellence in this area results into higher customer satisfaction leading to brand loyalty, and the indirect lift to future vehicle sales. We are looking at significant benefit ranges: inventory reduction up to 30% while maintaining or improving fill rates.
Auto OEMs need to ensure they are not left behind on the service revolution. A properly managed and optimized spare parts business can boost revenue, profitability, and brand image simultaneously. OEMs need to consider how to integrate the spare parts business with the overall strategic plan, how to provide end-to-end supply chain visibility, how to leverage this visibility into demand & inventory levels to optimize business performance through granular and accurate demand sensing & multi-echelon inventory optimization. OEMs need to evaluate the maturity levels of current ERP provider systems and assess how much money they are leaving on the table by not taking advantage of true Integrated Business Planning to optimally balance cost, service, and profitability and evaluate multiple trade-offs through what-if scenario management. We believe forecast error can be reduced by 40-50%, inventory levels can be reduced by 20-30%, and net profit can be improved by 2-5% of revenue through use of advanced analytics in the spares supply chains.
Ashutosh Bansal is the Founder & CEO of GitaCloud.
Incorporated in Delaware, GitaCloud is on a mission to improve integrated business planning and decision making competencies at it clients. GitaCloud Principals come from a rich background of helping dozens of leading Fortune 500 companies through their business transformation in Sales & Operations Planning, Demand Planning, Supply Chain Planning & Optimization domains. GitaCloud offers a full range of services: business transformation advisory, reselling best of breed cloud software, Systems Integration engagements, and Supply Chain Managed Services. GitaCloud clients range from Automotive, High-Tech, Pharmaceutical, Consumer Goods, and Retail verticals across North America and Asia Pacific markets. For more information, please visit www.GitaCloud.com.