Inventory optimization applies to the resolution of two types of problems: what inventory objective should be had in each warehouse and in the case of multilevel networks, where said inventory should be located.

Without inventory optimization, companies typically set inventory goals based on simple rules or calculations made at a single level. The former normally define the inventory objective in days of coverage. In single-level calculations, you examine a single item at a single location and calculate the amount of inventory needed to meet demand.[9]​.

Inventory optimization models can be deterministic (in which each set of variables is uniquely determined by the model parameters) or stochastic (in which variables are described by probability distributions).[10]​ Stochastic optimization takes into account supply uncertainty, so that, for example, 6% of orders from a foreign supplier have a delay of 1 to 3 days, 1% have a delay of 4 to 6 days, 5% have a delay of 7 to 14 days and 8% have a delay of more than 14 days.[11]​.

Stochastic optimization also takes into account demand volatility, which is one of the main challenges that supply chain professionals face.[12]​ Instead of considering demand forecasting as represented by a single number, stochastic models estimate that, for example, there is a 65% probability of selling 500 units, a 20% probability of selling 400 units, and a 15% probability of selling 600 units. To achieve a very high level of service, an alternative is to have a lot of inventory and have a very reactive transportation system, which will lead to a high logistics cost. But high service levels can be achieved if costs increase, emergencies to be managed and if there is too much inventory; But greater profitability can be obtained if the origin of the volatility is known and adequate planning is carried out. The result of the stochastic approach is that it allows for a better understanding of inventory requirements than that provided by a deterministic approach.[11]​.

A single-level sequential approach predicts demand and determines the inventory needed for each level separately. Multi-level inventory optimization calculates the correct inventory levels across the network based on demand variability at various nodes and performance (lead time, delays, and service level) at higher levels of the network.[13]​.

Multi-level inventory optimization examines inventory levels globally across the entire supply chain while taking into account the impact of inventories at any level on other levels. For example, if the product sold at a retailer's point of sale is received from one of its distribution centers, that center represents one level of the supply chain and the point of sale represents another. It should be clear that the amount of stock needed at points of sale is a function of the service received from the distribution center. The better the service provided at higher levels of the supply chain, the less inventory needed at lower links. The goal of multilevel inventory optimization is the continuous determination of safety stock values ​​at all network levels.[4]​.

Multi-level inventory optimization represents the next-generation approach to optimizing inventory across the entire supply chain. Modeling the different levels of the chain allows the accurate prediction of other types of inventory, such as cycle stock and safety stock. As part of inventory optimization, supplier performance, customer service, and internal asset performance metrics should be constantly monitored to enable continuous improvement.[14]​.

Several companies have obtained financial benefits from the application of this inventory optimization approach. A study carried out by IDC Manufacturing Insights has shown that many organizations that have applied inventory optimization have managed to reduce inventory levels by up to 25% in one year and have achieved cash flow of more than 50% in less than two years.[6]​

Electrocomponents"), one of the world's largest distributors of electronics and maintenance products based in the United Kingdom, has increased its profits by £36 million by optimizing inventory while simultaneously achieving higher service levels with less inventory. line fulfillment).[16]​ Smiths Medical, a division of Smiths Group, used inventory optimization to better address demand volatility and supply variability, reducing both the risk of overstocks and stockouts, while optimizing the manufacturing process.[17]​.

Inventory optimization applies to the resolution of two types of problems: what inventory objective should be had in each warehouse and in the case of multilevel networks, where said inventory should be located.

Without inventory optimization, companies typically set inventory goals based on simple rules or calculations made at a single level. The former normally define the inventory objective in days of coverage. In single-level calculations, you examine a single item at a single location and calculate the amount of inventory needed to meet demand.[9]​.

Inventory optimization models can be deterministic (in which each set of variables is uniquely determined by the model parameters) or stochastic (in which variables are described by probability distributions).[10]​ Stochastic optimization takes into account supply uncertainty, so that, for example, 6% of orders from a foreign supplier have a delay of 1 to 3 days, 1% have a delay of 4 to 6 days, 5% have a delay of 7 to 14 days and 8% have a delay of more than 14 days.[11]​.

Stochastic optimization also takes into account demand volatility, which is one of the main challenges that supply chain professionals face.[12]​ Instead of considering demand forecasting as represented by a single number, stochastic models estimate that, for example, there is a 65% probability of selling 500 units, a 20% probability of selling 400 units, and a 15% probability of selling 600 units. To achieve a very high level of service, an alternative is to have a lot of inventory and have a very reactive transportation system, which will lead to a high logistics cost. But high service levels can be achieved if costs increase, emergencies to be managed and if there is too much inventory; But greater profitability can be obtained if the origin of the volatility is known and adequate planning is carried out. The result of the stochastic approach is that it allows for a better understanding of inventory requirements than that provided by a deterministic approach.[11]​.

A single-level sequential approach predicts demand and determines the inventory needed for each level separately. Multi-level inventory optimization calculates the correct inventory levels across the network based on demand variability at various nodes and performance (lead time, delays, and service level) at higher levels of the network.[13]​.

Multi-level inventory optimization examines inventory levels globally across the entire supply chain while taking into account the impact of inventories at any level on other levels. For example, if the product sold at a retailer's point of sale is received from one of its distribution centers, that center represents one level of the supply chain and the point of sale represents another. It should be clear that the amount of stock needed at points of sale is a function of the service received from the distribution center. The better the service provided at higher levels of the supply chain, the less inventory needed at lower links. The goal of multilevel inventory optimization is the continuous determination of safety stock values ​​at all network levels.[4]​.

Multi-level inventory optimization represents the next-generation approach to optimizing inventory across the entire supply chain. Modeling the different levels of the chain allows the accurate prediction of other types of inventory, such as cycle stock and safety stock. As part of inventory optimization, supplier performance, customer service, and internal asset performance metrics should be constantly monitored to enable continuous improvement.[14]​.

Several companies have obtained financial benefits from the application of this inventory optimization approach. A study carried out by IDC Manufacturing Insights has shown that many organizations that have applied inventory optimization have managed to reduce inventory levels by up to 25% in one year and have achieved cash flow of more than 50% in less than two years.[6]​

Electrocomponents"), one of the world's largest distributors of electronics and maintenance products based in the United Kingdom, has increased its profits by £36 million by optimizing inventory while simultaneously achieving higher service levels with less inventory. line fulfillment).[16]​ Smiths Medical, a division of Smiths Group, used inventory optimization to better address demand volatility and supply variability, reducing both the risk of overstocks and stockouts, while optimizing the manufacturing process.[17]​.