Distribution Experience
I ran a project for a specialist European carrier of Carpets: its business is in collecting cut carpet from manufacturers, bringing it into its distribution depots and from there sending it out to retailers.
The business sought simply to barcode carpet so it was easier to track in and out of depots. My involvement eventually brought about a major re-design of the way depots operate as well as significant changes in how staff were incented and rewarded. My changes saved the business millions, and continue to show a return several years on.
In this engagement, we needed to:
- Reduce critical staff dependencies
- Speed parcels through our system so as to reduce overheads
- Reduce damage to and loss of parcels
- Realise efficiencies through automated route planning
- Increase customer satisfaction, to inoculate ourselves against predatory competition
- Prevent the system grinding to a halt in the fourth quarter
and I added:
- Keep our people on-side
Barcodes, Pods and Spikes
Bar-coding carpet made it possible for the first time to track carpet through a depot and to know exactly where every piece is at all times. This was a critically important innovation: previously, carpet would be grouped by destination postcode and needed re-picking from that code’s bay. This was a remarkably skilled job — as I discovered when I tried to identify the roll of 4.2m of moss green Axminster that I needed — just from its end profile.
The year’s peak for carpet purchase is in October and November — that’s after summer, but as spare rooms are being smartened up for Christmas. So when depots were full to bursting and the first of the winter bugs were keeping some of the skilled staff at home we could not turn to casual labour to support the picking work: they’d be only as much use as I had been.
So it was critical to remove this dependency from the way depots worked, and this was achieved by placing carpet in identified addresses (or ‘pods’) from which anyone could pick: now you were looking for position K-81; not that particular shade of green. This approach also had other knock-on effects for depots. One was that hotspots became a thing of the past as postcode clustering was no longer necessary. And when you realise that carpet is picked and moved with trucks that have four-metre poles at the front which pick carpet up by spiking the centre tube, it becomes obvious that flow around depots is important for both efficiency and safety.
The business had been looking for additional depot space and even an additional regional depot. But once the depots were working efficiently the capacity needed was dramatically reduced. Rather than move into new space, we were able to close and relinquish a number of depot annexes.
Delivery Routes
We next looked at automated route planning. This was a challenge to established drivers who simply ‘knew’ their patch and would pick, load and deliver themselves using their own specialist knowledge. The pick order of carpet is important: if it’s not picked in reverse order of delivery, you could end up with a required piece of carpet that’s underneath all the others and irretrievable.
Automated route planning was initially a disaster. It couldn’t call on a driver’s expert knowledge, so it had no clue that there was an immutable rule that ‘drop seven can only be in the afternoon, as the shop is only staffed by one person until two o’clock’. It was only by allowing that information to be factored in, and by making the routes ‘advisory only’ that we slowly made the route-defining process better and ultimately broadly used.
Wonky Tachographs
As I looked further into the efficiencies of the distribution process, I became perplexed by something that the tachographs appeared to be telling me: that our lorries were often-as-not parked up at 16:45. I knew this had to be wrong, but it took a long time and a lot of carpet-humping and credibility-building (with the lorry drivers) before I worked out what really was happening. And the tachos were right.
Drivers were delaying their return to depot because if they came back after 17:00, an automatic overtime payment became due. I found their behaviour entirely rational, but some Directors of the business were so affronted that they saw the issue as a disciplinary one.
I saw it as an example of bright people exploiting a daft process, so I insisted that we change the process, and made it attractive to return to the depot as soon as the round was finished. That meant we knew how long rounds really took, and could better balance them so they filled the working day. Again, huge savings — some very hard to measure — were realised.
But we most significantly gained by engaging staff as thinking adults rather than deployable work units: by giving staff opportunities and trust, they helped us run the business better. They made more money for themselves, spent less time idle and went home when they’d finished their work. Good result.