Managing slow-moving item: a zero-inflated truncated normal approach for modeling demand

Area of Science:

• Operations Research
• Inventory Management
• Statistical Modeling

Background:

• Managing inventory for slow-moving items with intermittent demand presents significant challenges in service enterprises.
• Traditional inventory models often struggle to accurately forecast and manage erratic demand patterns.
• Effective inventory control is crucial for minimizing costs and maximizing service levels.

Purpose of the Study:

• To propose an advanced inventory management method for slow-moving items in service enterprises.
• To develop a statistical distribution capable of modeling intermittent demand effectively.
• To enhance the performance of continuous review inventory models with shortages.

Main Methods:

• Utilized a zero-inflated truncated normal statistical distribution to model intermittent demand.
• Employed numerical experiments with an intermittent demand forecasting algorithm over fixed lead times.
• Applied the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) for multi-criteria decision analysis.

Main Results:

• The proposed statistical distribution significantly improved the performance of continuous review inventory models.
• The new method outperformed traditional approaches like simple exponential smoothing and Croston's method.
• A positive relationship was identified between demand intermittency variability and reorder point decisions.

Conclusions:

• The developed method offers an original and valuable approach for slow-moving item management in service companies.
• The statistical distribution accurately models intermittent demand, leading to better inventory control.
• The study provides a robust framework for multi-criteria decision verification in inventory management.