Inventory MGT in Uncertainty

MANAGING INVENTORY IN AN UNCERTAIN WORLD

Mike Dixon, PhD

Demand variability presents two fundamental challenges in operations management: variations in how much customers want (quantity variability) and when they want it (timing variability). A restaurant might face both challenges during a typical week- not only do different numbers of customers arrive each day, but they also cluster at different times, creating unpredictable peaks and valleys in demand. Similarly, a manufacturer might see seasonal spikes in order volumes while also experiencing random surges in demand throughout the year.

This variability creates a fundamental tension with Just-in-Time (JIT) principles. While JIT systems strive for minimal inventory levels and perfect synchronization with demand, real-world variability makes such perfect synchronization challenging. Runningwith extremely leaninventory levels, as advocated by JIT, leaves an organization vulnerable to disruptions whendemand patterns deviate from expectations.

This is where safety stock and buffer inventory emerge as strategic tools for managing variability while still maintaining relatively lean operations.

Safety Stock is the additional quantity of an item held in inventory beyond the expected demand during the lead time. It acts as insurance against demand spikes or delays in replenishment. The primary goal of safety stock is to prevent stockouts and maintain service levels without excessively increasing holding costs.

Safety Stock specifically addresses external variability- the uncertainty in customer demand and supplier performance. For instance, a retailer might maintain additional inventory to cover unexpected spikes in demand or delays in supplier deliveries. The amount of safety stock needed increases with both the variability of demand and the desired level of customer service.

More Variability in Demand and Supply = More Safety Stock

Buffer inventory, on the other hand, focuses on internal variability- the normal fluctuations in production rates, setup times, and process yields. While JIT principles push for the elimination of these buffers through process improvement, some level of buffer inventoryoften remains necessary to maintain smooth operations, especially inprocesses that haven’t achieved perfect stability.

More Variability in Processing = More Buffer Inventory

The key lies in finding the right balance. Too little safety stock or buffer inventory leaves an operation vulnerable to disruptions; too much defeats the purpose of lean operations and JIT principles. This balance depends on several factors:

• The predictability of demand (more variable demand requires more safety stock)
• The reliability of suppliers (less reliable suppliers necessitate more safety stock)
• The stability of internal processes (less stable processes need more buffer inventory)
• The cost of stockouts versus the cost of holding inventory
• The ability to respond quickly to changes in demand

Organizations often use data analysis to optimize these inventory levels, analyzing historical demand patterns and supply chain performance to set appropriate safety stock levels while continuously working to reduce the need for buffer inventory through process improvements.

Safety Stock Calculation

Several factors affect the calculation of safety stock:

1. Demand Variability: Fluctuations in customer demand during the lead time.

2. Lead Time Variability: Inconsistencies in the time it takes to receive replenishment orders.

3. Desired Service Level: The probability of not experiencing a stockout during the lead time. A higher service level requires more safety stock.

4. Forecast Accuracy: The precision of demand forecasts.

There are various methods to calculate safety stock, ranging from simple rules of thumb to more complex statistical models. Below is a common approach:

Basic Safety Stock Formula

A simple way to calculate safety stock is:

• Z-score: The number of standard deviations corresponding to the desired service level (from the standard normal distribution, see below for more description).
• =Standard deviation of demand.
• L: Lead time in periods.

Example:

• Desired service level: 95% (Z-score ≈ 1.65)
• Standard deviation of daily demand (=50 units)
• Lead time: 10 days

The result suggests maintaining a safety stock of 260 units to achieve a 95% service level with the given level of variability (standard dev of demand is 50 units). Holding less than that means you will have a risk of running out before receiving the next shipment.

Other formulas can include variability in lead time. Formulas can be found online with a simple search.

Z-Scores and Service Levels

A service level represents the probability of not running out of stock during a replenishment cycle. The Z-score is the statistical value that corresponds to this probability. Here’s how they connect:

Common Service Level to Z-Score Conversions:

• 84% service level = 1.00 Z-score
• 90% service level = 1.28 Z-score
• 95%service level = 1.96 Z-score
• 98% service level = 2.05 Z-score
• 99% service level = 2.33 Z-score
• 99.9% service level = 3.08 Z-score

In practical terms:

• A 95% service level means you expect to have stock available 95% of the time
• The higher the service level, the more safety stock is required
• Moving from 95% to 99% service level requires significantly more inventory
• The relationship isn’t linear- getting that last 1% of service level requires a disproportionate increase in inventory

Business Impact Example:

• Critical medical supplies might warrant a 99.9% service level
• Common retail items might use 95%
• Low-priority maintenance supplies might use 85%
• Each step up in service level increases inventory costs but reduces stockout risks

Inventory Control Systems: Continuous and Periodic Review

Effective inventory management requires not only calculating appropriate stock levels but also deciding how and when to monitor and reorder inventory. Two primary systems used for this purpose are continuous review and periodic review inventory management systems. Understanding these systems helps organizations choose the approach that best aligns with their operational needs and inventory characteristics.

Continuous Review System

Definition: A continuous review system, also known as a perpetual inventory system, involves constant monitoring of inventory levels. Each time an item is withdrawn from inventory(due to a sale, production use, or any other reason), the inventory records are updated in real-time. When the inventory level of an item falls to or below a predetermined reorder point (ROP), a fixed quantity is ordered to replenish the stock.

Key Characteristics

• Real-Time Monitoring: Inventory levels are tracked continuously, often using inventory management software or electronic point-of-sale (POS) systems.
• Fixed Order Quantity: The quantity ordered upon reaching the reorder point is typically constant, known as the Economic Order Quantity (EOQ) or a predetermined lot size.
• Reorder Point (ROP): The inventory level at which a new order is triggered. It accounts for lead time and safety stock.

Advantages

• Timely Replenishment: Continuous monitoring ensures that orders are placed precisely when needed, reducing the risk of stockouts.
• Lower Safety Stock Requirement: Since inventory is closely monitored, less safety stock may be needed compared to periodic systems.
• Better Visibility: Real-time data provides insights into inventory trends, enabling more accurate forecasting and decision-making.

Considerations

• Higher Monitoring Costs: Requires investment in technology and systems to track inventory continuously.
• Not Ideal for Low-Value Items: The cost of continuous monitoring may outweigh the benefits for items with low demand variability or low value.
• Complexity: May be more complex to manage, especially for organizations with a large number of items.

Suitable For

• Items with high value or critical importance.
• Inventory with unpredictable or highly variable demand.
• Organizations that can invest in inventory management technology.

Periodic Review System

Definition: A periodic review system involves checking and reviewing inventory levels at fixed intervals (e.g., weekly, monthly). At each review, the inventory position is assessed, and an order is placed to raise the inventory level up to a predetermined target level or order-up-to-level.

Key Characteristics

• Fixed Review Intervals: Inventory is checked and orders are placed at regular, scheduled times.
• Variable Order Quantity: The quantity ordered varies depending on the inventory level at the time of review.
• Simplified Monitoring: Inventory records may not be updated continuously; physical counts or estimates are used during reviews.

Advantages

• Simplified Operations: Easier to manage, especially for smaller organizations or those with limited resources.
• Lower Monitoring Costs: Does not require constant tracking, reducing administrative burden.
• Bulk Ordering Opportunities: Can align orders for multiple items, potentially reducing ordering costs through consolidated shipments.

Considerations

• Higher Safety Stock Requirement: Longer intervals between reviews increase the risk of stockouts, necessitating more safety stock.
• Less Responsive: Delays in detecting inventory depletion can lead to missed sales or production delays.
• Potential for Overstocking or Stockouts: Variability in demand between review periods can lead to inaccurate ordering.

Suitable For

• Items with stable and predictable demand.
• Low-value items where the cost of stockouts is low.
• Organizations with limited technological capabilities for continuous monitoring.

Comparing Continuous and Periodic Review Systems

ABC Inventory Classification

ABC classification guides the choice between continuous and periodic review systems. ‘A’ items, representing roughly 20% of inventory items but 80% of value or criticality, warrant continuous review to maintain tight control over these crucial assets. ‘B’ and ‘C’ items, being less critical or valuable, can be efficiently managed through periodic review, with longer intervals between checks for ‘C’ items than ‘B’ items. This differentiated approach optimizes resources by matching the intensity of monitoring to each item’s importance.

Technology in Inventory Management

RFID and Barcode Systems

Modern inventory tracking relies heavily on automatic identification technologies. Barcodes offer a cost-effective solution for item tracking, while RFID enables contactless scanning of multiple items simultaneously. RFID tags are particularly valuable in complex supply chains or when items need tracking through multiple locations, though they require a higher initial investment than barcode systems. Both technologies dramatically reduce human error in inventory counting and provide real-time visibility of stock movements.

Inventory Management Software and Enterprise Systems

Enterprise Resource Planning (ERP) and Material Requirements Planning (MRP) systems integrate inventory management with other business functions. These systems automate ordering, track stock levels, generate demand forecasts, and provide analytics for decision-making. While MRP focuses specifically on production planning and inventory control, ERP systems offer broader integration, connecting inventory management with accounting, sales, and other business operations. Modern cloud-based solutions make these tools accessible to businesses of all sizes, enabling data-driven inventory decisions and improved supply chain visibility.

Vendor-Managed Inventory (VMI)

Vendor-Managed Inventory shifts inventory management responsibility to suppliers, who monitor stock levels and make replenishment decisions on behalf of their customers. Under VMI agreements, suppliers access their customers’ inventory data and use this visibility to maintain agreed-upon stock levels, often through automated systems. This collaborative approach reduces administrative burden for buyers, minimizes stockouts, optimizes order quantities, and typically leads to better inventory turns and lower carrying costs for both parties. For example, many retail chains use VMI with major suppliers who manage their own product categories within stores