How shippers and carriers can leverage analytics to improve the performance of their private fleets using near-real-time shipment location data, combined with customer orders, routing plans, electronic proof-of-delivery data, service/work order schedules, vehicle inspection/maintenance data, and other sources of data.
Part Two: Analytics for Private Fleet and Driver Performance—Here In Part Two, we look at how analytics uses real-time location data, combined with orders, plans, proof-of-delivery, vehicle data, and more to drive significant improvements to fleet and driver performance.
Analytics for Improving Private Fleet and Driver Performance
Shippers and/or carriers operating their own fleets can benefit by using analytics to better understand and improve actual performance vs. plans and customer commitments, vehicle utilization and performance, driver performance, cost-to-serve, and more.
Precise nearly real-time data about vehicle locations is becoming more widely available due to the mandated adoption of Electronic Logging Devices for all commercial vehicles in the US and Canada and/or voluntary adoption of driver tracking apps. This location data can be combined with customer orders, routing plans, electronic proof-of-delivery data, service/work order schedules, vehicle inspection/maintenance data, and many other kinds and sources of data to provide all kinds of analytics to better understand and improve the performance of private fleets.
In this article and in Part 2B of this series, we look at four kinds of private fleet analytics that fleet owners are doing:
1) Overall Fleet Performance to Plans and Commitments,
2) Vehicle utilization and performance,
3) Driver performance, and
4) Cost-of-service tradeoffs.
Overall Fleet Performance to Plans and Commitments
Analytics can help measure and improve performance vs. the plan and/or commitments made to the customer. This can be broadly grouped into two areas: 1) improving overall performance and 2) improving plan accuracy/optimization effectiveness:
Improving overall performance—Analytics can be used to identify where the fleet is falling short in end-to-end performance, such as on-time delivery and exception-/damage-free delivery. Analytics can help isolate the issues, such as problems with performance from a particular facility, driver, vehicle, time-of-day, or route/customer. More importantly, by drilling down on the details, analytics can help answer why performance is suffering so that more effective mitigation strategies can be formulated and put into action. Dashboards can help track progress against improvement goals and provide KPIs for individuals, groups, and facilities to better understand how they stack up.
Improving plan accuracy and optimization effectiveness—Transportation/delivery plans tend to be based on averages and/or estimates, rather than taking into account differences between sites, drivers, products, time-of-day, and so forth. As a result, individual legs and stops within a plan, as well as the overall plan, will be too conservative (resulting in underutilization of resources), or too optimistic (resulting in missed commitments, disappointed customers, and frustrated employees, including potentially overly aggressive driving to compensate for a too-tight schedule). Some of the factors that analytics can measure to adjust transit and service times include:
Customer site characteristics—Actual delivery time is impacted by site characteristics such as the security check-in process, size of facility and floor of delivery, availability and speed of elevators, customer’s receiving process and receiving staff availability, and so forth. Specific data on these may or may not be available. Lacking those data, GPS data can be used to record the actual time-on-site, in order to more accurately estimate the time allotted for future deliveries to that same location. Analytics can help detect if time-on-site changes based on time-of-year, or when a trending or sudden change to average time-on-site has occurred. Geocoding can be made more accurate by recording the actual location of delivery.
Product/shipment characteristics—Size, weight, handleability, and special handling requirements (e.g. delicate or hazardous) all impact the delivery time, especially loading and unloading.
Driver differences—Differences in experience, physical fitness, and familiarity with the route and/or customer site can impact delivery times. Expectations can be adjusted as a driver gains experience.
Vehicle differences—Size and maneuverability may determine how close a vehicle can get to the point of delivery and thereby how far the delivery has to be carried by hand or cart—in some cases it might be hundreds of yards, making a material difference in delivery time. The availability and type of on-vehicle equipment/capabilities, such as a liftgate, crane, forklift, etc. can make a difference. On the road as well, the actual speed that different vehicles travel along various routes and under different conditions will vary based on the type of vehicle (e.g. semi vs. step van) and these differences can be derived via analytics using GPS and other relevant data.
Time-based differences—Time of season, day-of-week, time-of-day, holidays, and timing of special events, can all affect transit and delivery times.
All this more granular understanding of factors impacting actual performance vs. plan will only be useful if your route planning and optimization system is able to incorporate it and adjust plans accordingly. In some cases, the planning system may already have built-in ability to adjust plans based on some of these characteristics (such as type of vehicle or product characteristics). However, it is also good if the adjustment factors in your planning system are extendable, to allow any arbitrary attribute you provide to impact transit and service times, with a flexible set of algorithms for determining the weight or influence of each attribute. This might include the ability to run scripts, so you or the solution provider can create simple algorithms to make transit- or service-time adjustments based on a combination of factors (e.g. heavy, hard-to-handle load + no elevator + number of flights of stairs).
In Part 2B of this series, we examine how analytics can be used to improve vehicle utilization and driver performance, as well as better tune priorities in planning and optimization software.
To view other articles from this issue of the brief, click here.