Optrak
Optrak Parcel
Optrak Parcel is designed to bring new optimisation techniques to the parcels industry that make maximum use of driver knowledge, take full account of rush-hour speeds to obtain highly efficient and effective routes.
Optrak Parcel is a cloud-based solution that takes advantage of the calculation power of multi-core “Compute Nodes” for processing, and sophisticated analytics combined with address-level geographical information to predict delivery times per address at the driver level, to give reliable, accurate plans.
The Problem
Online shopping has seen extraordinary market growth in the last five years, and 2020 and COVID-19 has only expanded that growth further. This means that traditional postcode-based planning – where each driver is allocated a patch of postcodes – is breaking down as areas have to be constantly re-planned to keep up with increasing but fluctuating demand.
Parcel companies planning delivery rounds have long been faced with a choice: base the routes geographically, to maximise driver knowledge or use a route planning package that will create different trips into different areas every day.
Current Planning
Geographical ‘postcode’ area routing has been favoured for two main reasons:
- Drivers are clearly more efficient and productive, in terms of time per parcel, when working in areas they know well.
- The system is inherently easier to manage. Parcels are allocated to drivers by postcode, resulting in a straight-forward sortation methodology.
The main disadvantage of this method is that it responds poorly to fluctuations in demand – drivers have to make informal or formal swap arrangements with each other when they have too many parcels, relief drivers may be brought in to run inefficient “mop-up” routes and drivers may face unpredictable working hours and payment as the number of parcels goes up or down.
Traditional optimised routing systems offer the promise of greater efficiency through optimisation, but they cannot easily adapt to different levels of individual driver productivity in different geographical areas; the optimisation efficiency gains can easily be outweighed by the lost productivity of drivers having to work in unfamiliar territory.
Solving This with Optrak Parcel:
Accurate Models
Optrak Parcel is able to combine on PoD and tracking data with accurate geographical data (in the UK making use of the Ordnance Survey AddressBase Premium of over 25m deliverable addresses) to give per-driver estimates of the time taken to park and deliver each parcel.
For drive-time data we use the historical TomTom drive-time information that models rush-hour traffic down to 5-minute time intervals.
Taken together, these time estimates will give an unparalleled level of predictive power for how long trips take, allowing the optimisation system to make the most effective use of driver availability.
Novel Optimisation
Optrak Parcel utilises a sophisticated pipe-line of optimisation techniques to produce routes that make best use of driver experience – allocating work to drivers on the basis of familiarity. New drivers can be allocated training areas – for instance to replace drivers no longer on roster – their workload, gradually increased as they pick up knowledge and become more efficient.
Optimisation is run in the cloud on a network of high-capacity “Compute Nodes” – computers with 64-core capability, to carry out large scale parallel computation.
Real-Time Feedback
Once planned, Optrak Parcel real-time will be able to link to PoD and other tracking facilities to manage a full real-time picture of the day’s plan as it unfolds. This will enable logging and support “your driver is … away” style web site reporting. Future versions of the software will be able to link to real-time traffic data to further improve predictive performance.
Learning Functionality
Each day’s trips and PoD data feeds back into the system to update driver capability, delivery conditions at particular addresses and overall system performance.
Incorporating historical data into the future model (using past delivery performance to inform drop times for previous sites and locations) as well as driver feedback from PoD systems will help Optrak Parcel improve the accuracy of stated delivery times for planned trips, driving higher delivery success and customer service levels.
Analytics can also be used at a management level, alongside PoD data to scrutinise driver performance and familiarity with areas to identify high performers and improved practises.
Supported by Innovate UK
Innovate UK drives productivity and economic growth by supporting businesses to develop and realise the potential of new ideas.
We connect businesses to the partners, customers and investors that can help them turn ideas into commercially successful products and services and business growth.
We fund business and research collaborations to accelerate innovation and drive business investment into R&D. Our support is available to businesses across all economic sectors, value chains and UK regions.
Innovate UK is part of UK Research and Innovation.
For more information visit www.innovateuk.ukri.org
