Technology Blogs by SAP
Learn how to extend and personalize SAP applications. Follow the SAP technology blog for insights into SAP BTP, ABAP, SAP Analytics Cloud, SAP HANA, and more.
Showing results for 
Search instead for 
Did you mean: 

Companies with complex distribution networks or spare parts supply chains are continually trying to achieve the balance between not enough and too much when it comes to inventories. Too little adversely affects service capabilities, while conversely, too much may result in costly overstocking and excessive storage.

To help alleviate these issues, customers can set up materials requirements planning (MRP) algorithms in their SAP ERP application, which order parts automatically based on the requirements that are visible at a point in time. The MRP algorithms use parameters such as safety stock, maximum level, and minimum order quantity in order to deal with the unavoidable uncertainties about future demand quantities and supply lead times.

The operation of a typical supply chain involves a large variety of materials, events, interactions and tactical decisions. Different types of demand with different degrees and types of uncertainty have to be served from the same stock. Stock locations are organized on multiple tiers that supply one another, so the stock availability in one location influences the supply lead time for another one. All this complexity forces the business to make limiting assumptions when looking for the best strategies and parameters, without having a good way to assess the real impact of these assumptions on the accuracy of the result. A lot of times the assumptions that are correct for a material might be completely off for another one. And sometimes, the real case of inventory problems is in a place so unlikely that no one is looking at it – the wrong number in a rarely used field, a mismatch between the units of measurement used in different locations, etc.

Unveil the real problems behind low service levels

Now there’s a way to have better insights into inventory control through simulation and optimization functionality developed by SAP Data Sciences and the custom development team.

This new functionality, which runs on the SAP HANA database, is an analytic “side-car” to solutions like SAP ERP and models a company’s actual supply chain, including all the interactions and events that impact inventory levels.

With this software, the expected future behaviour of the full supply chain is simulated in detail, revealing all interactions and allowing for the identification of known and unknown problem areas. Based on the outcome of the simulation, a company’s current strategies and parameters in the supply chain can be adjusted to ones that produce an ideal outcome.

Realistic simulation, real-life optimization

The simulation model includes both deterministic events (such as purchases and stock transfers) as well as stochastic events (unexpected demand and lead-time variability). The stochastic events are described by statistical models which are custom built from the historical data and enhanced using the experience of the business users.

Starting from the current state of the supply chain and including the existing MRP algorithms, the simulation computes the future inventory situation in detail.
The optimization algorithms make use of the simulation to find the ideal replenishment strategies and parameters.

Empower supply chain analysts

The interface of this functionality brings together all the information about the current state, the expected future state and the optimal state of the supply chain. Supply chain analysts are given a complete view of their supply chain – they can view KPIs that reflect all materials, in all locations, over a long period of time and quickly switch to a view of the expected stock level or demand for one material, in one location, on a daily basis.

They can simulate the effects of any parameter changes and even new configurations of the network and in this way get additional insight before implementing changes in the real system.

This functionality allows companies to:

  • Visualize consolidated information for each material part across the whole supply chain
  • Predict the behaviour of the supply chain in the future, based on a data snapshot from SAP ERP
  • Simulate the effect of the existing network configuration, replenishment strategies, and parameters
  • Calculate KPIs, such as service levels or average inventory, based on the simulations
  • Optimize inventory policies to achieve the best possible trade-offs between service level and inventory stocking costs

The key: Event level simulation of the supply chain

At the heart of this new functionality is the simulation of events on the supply chain, on a daily level.

Data - such as weekly historical demand, future planned demand, MRP parameters, supplier and internal lead times - can be pulled from SAP ERP, other applications and even flat files and brought together in SAP HANA to form a coherent, detailed and easily accessible supply chain overview.
Rules – such as the quantities that MRP would order or priority of some types of demand – are modelled by a simulation algorithm written in JAVA, which allows the fast simulation of the effect of a large number of individual events on the situation of the supply chain.

The simulation algorithm is able to compute what will happen on the supply chain in the future and report the possible ranges for all supply chain KPIs: average stock values, order values, numbers of orders, average waiting days, etc.

The optimization algorithms are able to modify all parameters a business user is able to modify in reality, and do so automatically until the expected simulation KPIs reach their optimal values.

The results of simulations and optimization are stored in SAP HANA along with the supply chain data that was copied from ERP and other sources. This, along with the next generation of user interfaces that the SAP HANA Platform enables, makes it possible for supply chain analysts to get a clear picture of the whole supply chain and the way in which present decisions influence future KPIs.

Improved service – and significant savings

With rapid simulation and optimization, the real problems behind low service levels become apparent and inventory issues are resolved.

And based on our experience of working with customers in this area, there is potential for significant savings by reducing average stock levels or order frequencies and increasing the service levels for materials which cause the most costly delays.

If you are interested in utilizing this technology, contact me to learn how this functionality can be integrated into your distribution network.

1 Comment