ResearchPicture source:


Industrial Engineer – Volume 46 Number 2

The right question for inventory management

Many manufacturing companies ask themselves if they should produce their products to stock or if they should start and order only when the respective customer and the due date are known. The advanced information processing technologies and the available features of top ERP systems lead to a situation in which orders arrive in a rather stochastic manner with very different times until their due dates. The manufacturer wants to be able to provide good service and meet due dates but also minimize waste and excess inventory costs. The variability in the due dates of arriving orders creates uncertainty that complicates the traditional MTO (make-to-order) versus MTS (make-to-stock) decision.

Professors Klaus Altendorfer and Stefan Minner identiff that making the right decision for MTO or MTS has a big influence on optimal costs. They show that the overall optimal minimum cost solution is reached when the available customer order information is use, as in and MTO system, and also some safety stock is included. This safety stock is equivalent to the base stock in MTS systems. And the authors also show that the right question should be: “How much safety stock should be applied for each single item being produced to order?”

Is the traditional warehouse best for order picking?

Order picking operations constitute the costliest activities in a warehouse, accounting for more than half of warehouse operating costs. A further analysis of order picking time reveals that around half of the time is spent travelling. This underscores the importance of utilizing the right layouts and routing strategies for order pickers to ensure overall efficiency of the warehouse.

One way to decrease travel time for order picking is to design the warehouse layout for efficient travel. A recent trend to achieve this end is applying “fishbone” layout, in which the middle aisles are angled instead of being parallel to the cross aisles of the warehouse. For single- and dual-command operations, this type of layout has been observed to decrease expected travel times by more than 20 percent and 10 percent, respectively.

In “Order Picking under Random and Turnover-Based Storage Policies in Fishbone Aisle Warehouses,” Melih Celik, a Ph.D. candidate from Georgia Institute of Technology, and Haldun Sural from Middle East Technical University, investigate how a fishbone layout compares to its traditional counterpart when multi-item pick lists are used.

The authors show travel time minimization on a fishbone layout can be solved optimally by transforming the warehouse layout graph to that of an equivalent traditional warehouse. Under random storage and uniform demand, a fishbone layout can perform up to around 30 percent worse than an equivalent traditional layout as the size of the pick list increases.  However, when demand is highly skewed and the warehouse uses turnover-based storage, a fishbone designs significantly outperforms traditional design regardless of the pick list size.