A simple heuristic policy for stochastic inventory systems with both minimum and maximum order quantity requirements

Area of Science:

• Operations Research
• Supply Chain Management
• Inventory Control

Background:

• Addresses limitations in existing inventory models for systems with minimum and maximum order quantity constraints.
• Highlights the complexity of determining optimal policies for such systems.

Purpose of the Study:

• To propose and evaluate a novel heuristic policy for periodic-review stochastic inventory systems with minimum and maximum order quantity requirements.
• To develop a method for computing system-wide long-run average costs and optimizing policy parameters.

Main Methods:

• Introduced a modified (s, S) heuristic policy.
• Employed a discrete-time Markov chain approach for cost computation.
• Developed an efficient algorithm for policy optimization.
• Provided bounds for optimal policy parameters (s and S).

Main Results:

• The modified (s, S) policy demonstrates excellent performance in numerical studies.
• The proposed method effectively computes long-run average costs.
• The study quantifies the impact of key inventory parameters on system performance.

Conclusions:

• The modified (s, S) policy is a practical and effective solution for inventory systems with MinOQ and MaxOQ constraints.
• The developed methodology provides valuable insights for optimizing inventory management in complex scenarios.
• The heuristic approach offers a computationally efficient alternative to finding exact optimal policies.