Procurement and Inventory Control Mechanisms for Critical Raw Materials in Semiconductor Supply Chains

The semiconductor industry has been regarded as a foundational pillar of global technological advancement, with widespread applications across computing, telecommunications, and the automotive sector [1-3]. Within the semiconductor supply chain, the procurement and inventory management of critical raw materials have remained pressing challenges that require immediate resolution. As semiconductor technologies continue to evolve and market demands undergo rapid transformation, the stable supply and rational inventory control of raw materials have become increasingly vital [4-6]. In response to these dynamics, the present study seeks to explore the mechanisms for the procurement and inventory control of critical raw materials in the semiconductor supply chain. The primary objective is to enhance the resilience and efficiency of the supply chain through scientifically informed procurement decisions and inventory management strategies. This approach is intended to support optimal resource allocation and strengthen market competitiveness in the context of growing global supply chain uncertainties.

Existing research on procurement and inventory management in semiconductor supply chains has primarily concentrated on the development of quantitative models and the application of optimization algorithms. Conventional methodologies have frequently relied on inventory control models driven by demand forecasting to minimize inventory costs and improve inventory management [7, 8]. However, such methods have often proven inadequate when confronted with the complex market environment and uncertainties inherent in real-world scenarios. First, many prior studies have assumed demand to be known and stable, despite the fact that demand within the semiconductor market is highly volatile and inherently difficult to predict [9-12]. Second, the majority of existing models have focused on the procurement and inventory management of single-material categories, neglecting the integrated handling of multiple types of critical raw materials. This narrow scope has limited the applicability of such models, particularly when addressing cross-industry demands and technological shifts [13, 14]. Furthermore, insufficient attention has been paid to the multi-tiered and multi-nodal complexity and uncertainty of the supply chain, including supplier management, global logistics, and unexpected supply disruptions [15-19]. As a result, current research approaches have exhibited notable limitations in practice, falling short of addressing the dynamic demands of the semiconductor market and the complexity of globalized supply chains.

This study is structured around two core components concerning the procurement and inventory control mechanisms of critical raw materials within the semiconductor supply chain. First, the procurement of critical raw materials in the semiconductor supply chain was examined. Emphasis was placed on how stable material supply can be secured on a global scale through optimized procurement strategies and effective supplier management. Key factors such as procurement diversification, long-term supply agreements, and supplier risk assessment were investigated. Quantitative analytical methods were employed to optimize procurement decisions, thereby addressing market uncertainty and volatility. Second, strategic inventory control mechanisms tailored to the semiconductor supply chain were proposed. Taking into account the industry-specific characteristics-particularly the high value and elevated risk of prolonged material overstock-inventory control strategies were developed for critical raw materials. Data-driven forecasting methods and flexible inventory control approaches were integrated to optimize inventory turnover and cost efficiency, while enhancing the responsiveness and adaptability of the supply chain. The present research addresses a gap in the existing literature by providing an integrated analysis of procurement and inventory management within the semiconductor supply chain. In addition to enriching the theoretical understanding of supply chain management, actionable strategic recommendations are expected to be derived for industrial application. By incorporating both procurement and inventory control dimensions, the study aims to strengthen supply chain resilience in semiconductor enterprises operating in increasingly complex market conditions. Furthermore, theoretical and practical insights are anticipated to support efficient and cost-effective global supply chain operations. The methodologies and conclusions presented may also offer valuable reference for supply chain management practices in other high-technology industries.

3.1 Overview of strategic inventory control methods in the semiconductor supply chain

Within the semiconductor supply chain, the primary objective of strategic inventory control is to ensure the sufficient availability of critical raw materials and semi-finished products throughout the production process while optimizing inventory levels to minimize costs and mitigate associated risks. Due to the extended production cycles, technological complexity, and high demand volatility characteristic of the semiconductor industry, considerable challenges have been posed to inventory management. To address the varying nature of independent demand and dependent demand, differentiated inventory control strategies have typically been adopted by semiconductor enterprises. In the case of independently demanded materials—such as finished products or certain critical electronic components—demand is highly uncertain and often difficult to forecast. In response to this uncertainty, inventory control methods based on demand forecasting and quantitative analysis have been implemented. These include models such as the Economic Order Quantity (EOQ) model and dynamic inventory optimization models, which aim to fulfill production requirements while minimizing inventory overstock and capital immobilization.

For inventory associated with dependent demand, more precise control methods have been employed. Such demand is typically derivable from product structure and specific ratios in the production process. For instance, in semiconductor manufacturing, the quantity of certain raw materials or components required is directly linked to the production schedule and material consumption rates of the primary product. Accordingly, inventory management strategies based on Material Requirements Planning (MRP) and Manufacturing Resource Planning (MRPII) systems have been utilized. These systems enable the automatic calculation of optimal procurement quantities and inventory replenishment levels for various raw materials and components, based on the progression of primary product manufacturing and associated material usage.

3.2 Strategic inventory control of independent demand in the semiconductor supply chain

Within the framework of strategic inventory control in the semiconductor supply chain, independent demand inventory models are generally categorized into two types: continuous and periodic review inventory systems, each exhibiting distinct applications within the semiconductor industry. The continuous review inventory system is particularly suited for raw materials and components characterized by high demand variability and extended production lead times. In semiconductor manufacturing, key components such as chips, integrated circuits, and other high-precision elements often experience significant fluctuations in demand and therefore require continuous monitoring and management. In such cases, real-time inventory tracking must be implemented, and the appropriate reorder point (EPO) must be established. Once inventory levels drop below the predefined reorder threshold, a replenishment order is triggered immediately. This system minimizes the risk of stockouts and is especially critical for high-value components with long production cycles.

In contrast, the periodic review inventory system is applicable to raw materials and components with relatively stable demand patterns and predictable supply cycles. In the semiconductor industry, this includes materials such as base chemicals and standardized electronic components, whose consumption remains consistent over time. Under this system, inventory levels are reviewed at fixed intervals—such as monthly or quarterly—to determine whether replenishment is required. This model is typically applied to non-core materials used in production processes, and the replenishment decisions for these materials are made at fixed time intervals, such as monthly or quarterly. Periodic review allows enterprises to maintain adequate inventory levels without requiring continuous monitoring, thereby reducing complexity and capital lock-in while maintaining overall supply chain stability. For the semiconductor supply chain, this approach facilitates efficient inventory management and helps reduce inventory control complexity and costs.

ping_mu_jie_tu_2025-06-04_105724.png

Figure 4. Fixed-order quantity model for strategic inventory control of independent demand in the semiconductor supply chain

The quantitative reorder method is particularly well-suited for critical semiconductor components that require precise inventory control and strict adherence to replenishment timelines. The core principle of this method lies in establishing a fixed EPO and order quantity based on known demand patterns and supply conditions. When the inventory level declines to the predetermined EPO, a replenishment process is automatically triggered. This ensures continuous availability of inventory and prevents production delays or disruptions in the supply chain caused by stockouts. The corresponding model is illustrated in Figure 4. This method typically involves two key steps:

a) Determination of the EPO

In semiconductor manufacturing, the demand for materials such as wafers and chip packaging components is often subject to fluctuation. As a result, it is essential to initiate replenishment once inventory levels drop to a specified threshold. The EPO is calculated by considering both the average daily demand and the lead time required for replenishment. Let W denote the order quantity per cycle and M denote the lead time. The basic formula for calculating the EPO is expressed as:

$EPO=E\times M$                 (1)

In practical applications, the impact of safety stock must also be considered, particularly in the presence of demand variability or potential delays in supply. When this factor is included, the EPO calculation is modified to incorporate the safety inventory level T, ensuring that production requirements can still be met during unforeseen supply disruptions. For instance, semiconductor companies may rely on historical data or predictive models to estimate the daily demand for each critical raw material. By multiplying the estimated daily demand by the projected lead time, the EPO can be determined. Once inventory falls below this threshold, the system is configured to automatically trigger a replenishment order, thereby preventing interruptions in production during the lead time period.

$EPO=EO+T$               (2)

b) Determination of the order quantity (W)

The order quantity (W) refers to the amount of inventory to be procured during each replenishment cycle. In the semiconductor industry, this is typically determined using the EOQ model. The core principle of the EOQ model is to minimize total inventory cost by balancing three primary components: purchasing cost, ordering cost, and holding cost. The objective is to identify an optimal order quantity that satisfies production requirements while avoiding excess inventory accumulation. Several factors must be considered in calculating the optimal order quantity W. First, the annual demand must be estimated, which is typically derived from market forecasts, production schedules, and sales projections. Second, the ordering cost per transaction and the annual holding cost per unit must be quantified, usually based on the firm’s operational expenditure structure. By applying the EOQ formula, the optimal order quantity W can be computed. This facilitates inventory cost minimization while maintaining sufficient stock levels to support uninterrupted production. In the context of the semiconductor industry—where long production cycles and complex supply chains are the norm—an accurately determined order quantity can prevent excessive capital investment in inventory while sustaining the efficiency of the production line. Let SXZ denote the total inventory cost, N represent the unit value of the item, and Q denote the annual holding cost as a percentage of the unit value. Let E represent the annual demand, and T the fixed cost per order.

$SXZ=EN+\frac{1}{2}WNQ+\frac{E}{W}T$        (3)

To determine the minimum economic order quantity W, the economic order quantity is denoted as W₀, the ordering cost per order as T, the annual demand as E, and the annual holding cost per unit as G=NQ. By differentiating the function SXZ with respect to W, the economic order quantity can be derived using the following expression:

${{W}_{0}}=\sqrt{\frac{2TE}{G}}$        (4)

The periodic review model is suitable for raw materials and semi-finished products that exhibit relatively stable and predictable demand patterns. Unlike the fixed-order quantity model, the periodic review model focuses on assessing inventory and initiating replenishment at regular, predefined intervals. This approach has been particularly effective in the semiconductor industry for managing items such as standardized electronic components and chemical inputs, where demand remains relatively constant. Its adoption reduces the complexity of inventory management.

The core principle of the periodic review model is to conduct inventory assessments at fixed time intervals (denoted as S), such as monthly or quarterly. At the end of each review period, the inventory balance is compared to a predetermined maximum inventory level (R). If the current inventory level (U) is below R, a replenishment order is issued to restore the inventory up to the target level. In this context, the order quantity W is calculated as the difference between R and U, such that W = R - U. Through this mechanism, semiconductor enterprises can maintain inventory within an optimal range—minimizing capital tie-up from overstocking while simultaneously preventing stockouts that could disrupt production schedules.

In practical applications, the periodic review model has been found to be particularly well-suited for semiconductor materials that are characterized by long production cycles and relatively stable inventory management requirements. For instance, certain foundational raw materials—such as standardized electronic components and basic chemicals—exhibit predictable consumption patterns in manufacturing and are typically associated with stable lead times. Based on historical consumption data and demand forecasting, appropriate review intervals (S) and maximum inventory levels (R) can be established to ensure that sufficient inventory is available at the end of each cycle to support uninterrupted production. Furthermore, because the periodic review model eliminates the need for continuous inventory tracking and instead relies on scheduled evaluations and replenishments, its management cost is relatively low. This feature renders it particularly effective for managing high-volume semiconductor materials with low demand variability. Specifically, let S denote the ordering interval. The economic ordering interval can be calculated using the following expression:

$SXZ=EN+\frac{T}{S}+\frac{EQNS}{2}$        (5)

The first derivative of the annual total cost with respect to S was set to zero, yielding the economic ordering interval:

${{S}_{0}}=\sqrt{\frac{2T}{ENQ}}=\sqrt{\frac{2T}{EG}}$        (6)

Additionally, the maximum inventory level (R) can be expressed as:

$R=E\left( S+M \right)$               (7)

3.3 Strategic Just-in-Time (JIT) inventory management in the semiconductor supply chain

Within the semiconductor supply chain, JIT inventory management serves as a highly effective control mechanism that significantly enhances supply chain efficiency and responsiveness. The core principle of JIT is demand-driven coordination, whereby inventory levels are minimized across all stages of the supply chain through precise demand planning and synchronization. The ultimate objective is to approach a “zero-inventory” state. This approach is particularly critical in the semiconductor industry, where production cycles are extended, technological advancements occur rapidly, and frequent production scheduling and supply chain coordination are required. The implementation of strategic JIT inventory management is not limited to inventory reduction alone. Rather, it involves the meticulous control of supply chain elements to reduce total operational costs while simultaneously increasing market responsiveness and improving supply chain agility.

Under the framework of strategic JIT management, every stage of the semiconductor supply chain is required to fulfill production demands at the right time, in the right quantity, and with minimal inventory. To achieve this, highly integrated relationships must be established between semiconductor manufacturers and their suppliers. These relationships rely on accurate demand forecasting and real-time information sharing to ensure that essential raw materials and components are delivered punctually and in appropriate quantities throughout each phase of the production process.

The production of semiconductor products is characterized by long lead times and multiple complex manufacturing processes, where any delay or inventory accumulation at a single stage may lead to production setbacks or full disruption of the overall schedule. To address this challenge, semiconductor enterprises must rely on real-time data exchange systems and highly accurate demand forecasting mechanisms to closely monitor market needs, production progress, and inventory status. At the core of JIT inventory management lies a pull-based control mechanism, wherein the production and delivery activities of upstream suppliers are triggered by the actual demand from downstream manufacturers. This approach ensures that overstocking and stockouts are simultaneously avoided. Through such a demand-driven replenishment mechanism, the consumption of materials during production is aligned precisely with supplier delivery schedules, enabling a substantial reduction in overall inventory levels throughout the manufacturing process.

The successful implementation of the JIT model in the semiconductor supply chain is contingent upon seamless information sharing and close collaboration among all stakeholders. Given the involvement of multiple suppliers and production units across stages—from raw material procurement through wafer fabrication, chip packaging, and testing, to final product delivery—tight coordination at each stage is imperative. To achieve this, enterprises are required to establish robust supply chain management systems that facilitate real-time data transmission and transparent information flow. These systems ensure that each stage of the supply chain is capable of responding rapidly to actual demand signals. This information-sharing infrastructure allows all participants in the supply chain to quickly understand demand fluctuations and production status, enabling timely operational adjustments. For example, critical raw materials used in semiconductor production—such as silicon wafers and chemical reagents—are subject to strict quality and delivery time requirements. Under the JIT framework, suppliers must be able to accurately monitor production progress and inventory needs, and schedule their own production and delivery accordingly. Real-time access to production plans enables suppliers to synchronize delivery timing and quantities with precision, thereby enhancing the responsiveness of the entire supply chain and preventing disruptions caused by delayed information or inaccurate forecasting.

3.4 Collaborative inventory control in the context of semiconductor supply chain management

The semiconductor industry is characterized by long production cycles, high demand volatility, rapid technological advancement, and frequent coordination across multiple enterprises and geographic regions. As such, inventory management within the semiconductor supply chain is no longer a matter of isolated stock control, but rather a comprehensive strategic issue involving demand forecasting, supplier relationship management, and logistics optimization. To enhance overall supply chain performance and remain competitive in an increasingly demanding global market, the adoption of advanced inventory control methodologies has become essential.

The core objective of inventory control in semiconductor supply chains is to optimize inventory levels, enhance responsiveness, and maximize inventory utilization efficiency, all while maintaining a high level of customer service and minimizing inventory-related costs. In this context, inventory is not merely viewed as stored goods but as a strategically managed resource that must be tightly aligned with dynamic demand signals and production cycle requirements. The overarching goal of strategic inventory management in this environment is to maximize supply chain effectiveness under conditions of uncertainty, ensuring that stockouts are avoided during production while simultaneously preventing excess inventory that may result in capital waste and increased storage costs. Specifically, this approach seeks to achieve the following:

a) Inventory reduction

Inventory levels are to be minimized as much as possible without compromising the ability to meet production and customer demand. This reduces warehousing and administrative costs.

b) Response speed enhancement

Through precise demand forecasting and dynamic inventory control, supply chain responsiveness must be improved. This ensures that fluctuations in market demand or production schedules can be swiftly addressed, thereby preventing supply chain disruptions caused by stock imbalances—whether shortages or surpluses.

c) Enhancement of customer service levels

Inventory control must ensure the ability to fulfill customer demand in real time, minimizing delivery delays or order losses caused by stockouts. This directly contributes to increased customer satisfaction.

d) Optimization of supply chain efficiency

Through collaboration and coordination across all stages of the supply chain, from raw material procurement to final product delivery, operational efficiency must be improved while maintaining lean inventory levels.

One advanced inventory control strategy that plays a critical role in the semiconductor supply chain is Vendor-Managed Inventory (VMI). Under the VMI model, inventory is managed directly by the supplier, who assumes responsibility for monitoring the customer’s inventory levels and proactively managing replenishment based on shared real-time data, including inventory status, production progress, and demand forecasts. This model effectively reduces the inventory management burden on semiconductor enterprises while enhancing supply chain efficiency. It is especially suitable for raw materials and components that exhibit high demand variability or involve long production lead times. For example, semiconductor manufacturing requires large volumes of precision components and chemical substances, both of which typically have extended lead times and impose significant logistical requirements. Through real-time monitoring of downstream inventory and production activity, suppliers operating under a VMI framework are able to accurately forecast demand and replenish inventory proactively, thereby avoiding production interruptions caused by material shortages. In addition, VMI allows for the centralization of procurement and transportation processes, leading to cost reductions, decreased inventory accumulation, and improved supply chain responsiveness. In the context of the semiconductor industry, VMI enables uninterrupted production while minimizing excess inventory, thus reducing capital lock-up and improving the overall capital efficiency of the supply chain.

Joint Managed Inventory (JMI) is an inventory control approach that emphasizes collaborative coordination among all nodes of the supply chain. By jointly formulating inventory plans, participating entities within the supply chain can achieve resource sharing and optimization, thereby enhancing overall supply chain efficiency. In the context of the semiconductor supply chain—characterized by numerous production stages and intricate interdependencies between suppliers and customers—independent inventory control often leads to the bullwhip effect, resulting in excessive or insufficient inventory across different stages. Through the synchronization of demand forecasting and inventory planning among supply chain partners, JMI provides an effective means of mitigating such imbalances. In practice, JMI is implemented through shared demand information, coordinated replenishment planning, and risk-sharing mechanisms. For instance, in semiconductor production, demand for certain raw materials and components may be highly variable. If each node in the supply chain acts independently, mismatches in supply and demand may occur, leading to bottlenecks or overstocking at specific stages. Under the JMI framework, joint decision-making guided by a central coordination mechanism allows all parties to align production planning with actual market demand. This ensures inventory levels are appropriately distributed across the entire supply chain, reducing variability and enhancing system-wide stability.

Both VMI and JMI serve as critical strategies within strategic inventory management for semiconductor supply chains. VMI enables suppliers to manage downstream inventory using real-time data acquisition and analytics, thereby reducing inventory levels while ensuring uninterrupted production. In contrast, JMI fosters collective participation and coordination across supply chain nodes, eliminating inventory fluctuations caused by information asymmetry or demand volatility. This joint approach optimizes overall inventory levels and mitigates the impact of localized stock issues on broader supply chain performance. A major challenge in strategic inventory management within the semiconductor sector lies in maintaining flexibility and stability in the face of globalization and intense market competition. To address this challenge, semiconductor enterprises are increasingly adopting advanced demand forecasting techniques, real-time data analytics platforms, and highly integrated supply chain management systems. These technologies enable rapid adjustment of inventory strategies in response to dynamic market conditions, thereby reducing risk while enhancing the agility and responsiveness of the supply chain.