Peapod has major distribution problems in New York, Het Financieele Dagblad reported last Friday. Ahold's US internet subsidiary has invested 60 million euros in a semi-automated distribution centre, but almost a year after start-up, the DC is still not delivering as promised. For example, the DC's designers had to go back to the drawing board twice. Conclusion: haste is seldom good.
A September 2013 article on SupplyChianDigest.com describes the struggle the DC's designers went through. That struggle starts with the sheer complexity of the operation. Peapod offers a wide range of products, ranging from animal feed and charcoal in bulk packaging to frozen foods that need to be stored at different temperatures. This means that a Peapod DC typically consists of nine different storage areas, each with its own assortment, order pattern, order picking characteristics and therefore, in the most extreme case, nine different storage and order picking systems. Since an order typically includes items from multiple storage areas, sooner or later the collected goods will also have to be consolidated. This leads to considerable challenges in planning and controlling the systems.
Low margins
Another challenge comes from the desire to automate as many processes as possible in the new distribution centre - on the one hand to minimise logistics costs and on the other to reduce lead times. The drive to minimise costs has everything to do with low margins. Competition is fierce, especially in New York, where Amazon has also launched an online supermarket. While Peapod has grown rapidly in recent years, it has never made a serious profit in the 25 years since it was founded, according to the FD. Striving for short lead times has everything to do with delivery times. Especially if same-day delivery becomes the standard, groceries must be ready for shipment within an hour or even sooner. That is virtually impossible in a fully manual DC.
No good prediction
What makes choosing the right storage and order picking systems with the right degree of automation difficult is the huge expansion of the assortment. To combat cut-throat competition, Peapod now offers thousands more items than before since the new DC was commissioned. The problem is that there was no good prediction of the demand for these new items, making it difficult to calculate the required capacity of the storage and order-picking systems.
Moreover, during the design phase, Peapod's logisticians and the hired consultancy had to go back to the drawing board twice. The first time because the budget for the new building was revised downwards by higher-ups, and the second time because the implementation costs turned out to be much higher than expected. When the budget was put together, no account had been taken of the fact that wages in the New York region are considerably higher than the US average.
Major concessions
In the end, Peapod opted for a semi-automated DC, set up by Dematic. For fast movers, the online web supermarket opted for zone picking with a combination of pick-to-light and barcode scanning, with the zones connected by conveyors. For medium movers, a traditional order picking process has been set up, except that the picking locations are automatically supplemented by a shuttle system. For the slow movers, a put-to-light system was implemented, with the goods being supplied via the same shuttle system.
Whether this combination of systems is optimal is difficult to judge from a distance here. What is clear is that Peapod made considerable concessions to the original design, especially as the total investment in hardware had to be revised downwards on two occasions. What is also clear is that Peapod was very ambitious when it stated in September 2013 that it wanted to go live within nine months. Eleven months later, the DC did indeed go live, but with the necessary problems it now appears.
Major risks
The way in which Peapod approached the project was far from ideal. Anyone seeking to set up such a complex distribution centre with so many different systems in such a short space of time is taking major risks. The chances of success would have been greater if the online supermarket had automated the system in phases rather than all at once, and had allowed more time for testing. The question is, however, whether online retailers such as Peapod have the time to do so.
As a result, Peapod could not deliver properly for a long time, which can be deadly in e-commerce. The Ahold subsidiary has been overtaken by Amazon in the New York region as the fastest and cheapest online supermarket and has had to revise its growth scenarios downwards.
Marcel te Lindert - Journalist logistics and supply chain
