Media Coverage
May 2010
What is the perfect warehouse?
If you run or own a warehouse I think you would be very interested in the answer to the question "What is the perfect warehouse?". Someone else who wanted to know this answer is Edward H. Frazelle, who is the director of the "Logistics Management Series" at Georgia Tech University in Atlanta Georgia. Georgia Tech has one of the most advanced supply chain and logistics departments in the world and their website is a very valuable resource for anyone interested in supply chain optimisation. The particular piece of research that I want to discuss in this week's column is the survey of warehouse practices conducted by Dr. Frazelle published here. This survey contains many very valuables insights into how some of the world's best companies are running the their warehouses and I want to mention just a few of these here.
The survey covers about 200 of the top companies in the US and each warehouse was ranked on its quality as follows. First the shipping accuracy (SA) was calculated as the percentage of order lines shipped without errors, then the inventory accuracy (IA) was calculated as the percentage of warehouse locations without discrepancies. Finally a warehouse quality index (WQI) was calculated as WQI=SA*IA. The warehouses were then ranked based on their WQI, to give a WQIR so that the best warehouse has rank of 1 and so on. Next a warehouse productivity rank (WPR) was calculated as the ratio of the annual number of order lines shipped to the annual total person-hours expended in the warehouse. Again these were ranked to give the WPRR and again the lower the WPRR the better the warehouse. Finally the productivity index (PI) was calculated as PI=WQIR+WPRR giving a balance between quality and performance. Again, the lower PI's are better.
Now there are some interesting results in the survey that correlate low PI's with different aspects of running a warehouse. The first one that I want to focus on is the impact of different practices on performance. The following practices were measured:
- Interleaving
- Activity Based Costing
- Directed Putaway
- Ergonomics
- Cross-Docking
- Slotting or Gliden Zoning
- Continuous Improvement Teams
- Advanced Shipment Notification
- Item Cube Information
- Cross Training
- Productivity or Time Standards
Then each warehouse was categorized based on the number of these practices that they had in place. This ranged from very few (0-3), few (4-6), many (7-9), very many (10-11) and then the average productvity index was calculated for the warehouses in each group. The average performances were as follows:
- Very few: 84
- Few: 82
- Many: 71
- Very many: 62
It is clear from this that better practices leads to better performance, but what is quite interesting is that there is not much difference in the performance from "Very few" to "Few", so it seems that you need to get quite a lot of your practices right before you see the benefit. There is another fascinating table comparing the degree of warehouse automation to the productivity index. This had the following results:
| Level of material handling automation | Productivity rank |
| Greatly above average | 59 |
| Above average | 49 |
| Average | 35 |
| Below average | 51 |
| Greatly below average | 50 |
It seems that in this case too much automation is a bad thing (remember that lower numbers are better). There are many more fascinating insights to be had and I would recommend downloading the survey and other resources available at the Georgia Tech website.
If you have any comments or questions, please e-mail me at david.lubinsky@opsi.co.za.
