Inventory Management: Types, Models, and Strategies
Types of Inventory
Understanding different types of inventory is crucial for effective inventory management:
- Cycle Inventory: Used to meet normal demand during a production cycle, replenished cyclically. Example: A bakery keeps a cycle inventory of flour, replenishing it weekly.
- Safety Inventory: Held to protect against uncertainties in demand and supply. Example: A hospital maintains a safety stock of essential medicines.
- Seasonal Inventory: Built up for predictable seasonal increases in demand. Example: A retailer stocks beachwear before summer.
Motivations for Keeping Inventory
Businesses maintain inventory for several strategic reasons:
- To Meet Demand: Ensure product availability for customer satisfaction. Example: A toy store stocks up before the holiday season.
- To Achieve Economies of Scale: Reduce per-unit costs by purchasing and producing in larger quantities. Example: A furniture manufacturer buys wood in bulk.
- To Hedge Against Price Fluctuations: Protect against supplier price increases. Example: An electronics manufacturer purchases extra components anticipating price rises.
Assumptions of the EOQ Model
The Economic Order Quantity (EOQ) model minimizes total holding and ordering costs, with these assumptions:
- Demand is known and constant.
- Lead time is fixed.
- Ordering and holding costs are known and consistent.
- No quantity discounts.
The EOQ model suits products with stable demand and predictable costs.
Assumptions of the POQ Model
The Production Order Quantity (POQ) model, an extension of EOQ, is used when inventory is replenished gradually:
- Production and usage occur simultaneously.
- Demand and production rates are known and constant.
- Single product focus.
The POQ model is ideal for manufacturing settings with simultaneous production and consumption.
Continuous vs. Periodic Review Systems
Advantages of Continuous Review System
- Immediate Replenishment: Allows immediate action when stock levels fall below a reorder point.
- Better Response to Fluctuations: Reduces the likelihood of stockouts.
Disadvantages of Continuous Review System
- High Monitoring Costs: Requires sophisticated tracking systems.
- Complexity in Management: More complex due to continuous monitoring.
Advantages of Periodic Review System
- Simplified Ordering Process: Orders are placed at fixed intervals.
- Consolidation of Orders: Can consolidate orders, potentially reducing costs.
Disadvantages of Periodic Review System
- Higher Safety Stock Requirements: Higher safety stock may be needed.
- Less Responsive: Less flexible in adapting to changes.
Why Forecasts Are “Always Wrong”
Forecasts are often inaccurate because they predict the future based on historical data and assumptions, which are inherently uncertain. To provide a “good” forecast:
- Use a mix of quantitative and qualitative methods.
- Continuously update forecasts.
- Incorporate scenario planning.
Greater Forecast Errors for Upstream Companies
Upstream companies experience greater forecast errors due to the “bullwhip effect,” where small changes in consumer demand amplify variations upstream.
Accuracy vs. Bias in Forecasting
Accuracy refers to the closeness of forecasts to actual outcomes (average size of error). Bias indicates a systematic deviation from actual values (consistently higher or lower).
Difference Between MAD and MSE
- MAD: Measures average absolute errors.
- MSE: Measures the average of the squares of the errors, giving higher weight to larger errors.
Analysts might prefer MSE when larger errors have exponentially greater consequences.
Moving Average vs. Simple Exponential Smoothing
- Moving Average: Calculates the average of data points within a specific number of past periods, suitable for stable environments.
- Simple Exponential Smoothing: Gives more weight to recent observations, suitable for data with trends but without seasonality.
Effect of a Larger Alpha in Simple Exponential Smoothing
A larger alpha makes the forecast more responsive to changes but also more sensitive to random fluctuations.
Double Exponential Smoothing
Used when data exhibits trends but no seasonal variations. It incorporates two smoothing constants: one for the level and another for the trend.
Winter’s Method
Extends exponential smoothing to data with seasonality and trends, using three smoothing equations: level, trend, and seasonal component.
The Newsvendor Model and Its Assumptions
The newsvendor model optimizes inventory levels under uncertainty, determining the optimal order quantity to minimize excess inventory and stockout risks. Assumptions include:
- Uncertain demand with a known probability distribution.
- Ordering decision made before demand is realized.
- Single ordering opportunity.
- Unsold goods have salvage value; stockouts incur a cost.
Example: A retailer uses this model to decide how many winter jackets to order before the season.
Importance of Supply Chain Contracts
Supply chain contracts define terms between parties, managing risk and sharing benefits. Example: A “buy-back” agreement where a manufacturer buys back unsold goods from a retailer.
Double Marginalization and Its Resolution
Double marginalization occurs when each supply chain entity adds its markup, leading to suboptimal pricing. Resolved through vertical integration or strategic partnerships like revenue-sharing agreements.
Importance of Revenue Management in Operations
Revenue management maximizes revenue potential through strategic control of inventory and pricing. Example: Airlines adjust ticket prices in real-time based on demand and other factors.
Challenges in Designing a Revenue Management System
- Data Collection and Analysis: Requires accurate, real-time data and advanced analytics.
- Customer Perception Management: Dynamic pricing can lead to customer dissatisfaction.
- Complexity in Implementation: Integrating with existing IT infrastructure can be complex.
Example: A hotel struggles to implement a system that dynamically adjusts room prices due to demand fluctuations, events, and competitor pricing.