packout ratio

May 16, 2022

SCS Engineers Environmental Consulting and Contracting route Optimization
Using KPIs and existing technology has great potential for collection route optimization.

 

When municipalities’ collection routes run like well-oiled machines, trucks make money for them, translating to lower rates for taxpayers and streamlined, effective curbside service. But waste haulers typically lack two potentially powerful tools to gauge efficiencies and inform cost-saving decisions for the most robust collection route optimization.

What’s missing are formal analyses and standardized key performance indicators (KPIs) tailored for this specialized, complex niche.

Some industry experts look to close the gap. They are developing and leveraging KPIs and analyses, aiming for more and better data to drive productivity. They find the two tools work well together: KPIs provide a baseline to inform the comprehensive analysis.

“What we are doing is standardizing the evaluation of collections to quantify performance and outcomes better using more detailed parameters than before,” says Kevin Callen of Route Optimization Consultants. He works with SCS Engineers to improve cities’ collection operations.

New performance parameters fill in where the number of stops, tons collected, and time on route leave off. These are the beginnings of KPIs, but to really tell the story of how well collection routes run, metrics must go deeper to assess near-countless variables potentially impacting outcomes. The human variable –actions of customers, drivers, and helpers – is part of that.

“You can have all the data in the world and have the route worked out to run like a Swiss watch. But the human factor is a wild card,” says Josh Krumski of SCS Engineers.

 

KPIs Can Peer Into Human Behaviors

Analyzing collection operations through a KPI lens sometimes enables municipalities to understand better the drivers’ judgment calls and how they play out. And it gives insight to help prepare for unpredictable circumstances in this fast-moving, changing industry.

“Ultimately, we are trying to set up methodology and identify best practices to improve route operations as they grow and change. It’s a systematic way to monitor operations closely. To determine if collections are as productive as possible and identify problems and underlying causes if they fall short,” Callen says.

Krumski leverages multiple KPIs to help with his toughest charge: balancing costs and service quality in an industry with a tight profit margin.

“When you see what towns bill to collect waste and recycling, then consider operational costs, it’s clear that if they run behind a few hours a day, it eats into their budgets. Time to get out of the truck, open the corral, service the container, put it back, and close the corral starts to add up,” he says.

SCS Engineers use KPIs to gauge more than what happens at curbside stops, leaning on them to provide objective, big-picture insight to municipalities too busy to vet as they run their daily operations. Below are KPIs that the team finds best to help inform their collection route optimization projects.

 

Maximum Dumps

At the top of the list is the maximation of dumps, which is about loading trucks to full capacity while minimizing commute time to and from disposal sites.

Crews should optimally do three dumps during a typical 10-hour route and two during an average eight-hour route. If they aren’t achieving this, the question is, how might they be able to?

 

Packout Ratio

Look for the answer in the packout ratio. This KPI defines the weight of waste in the truck versus the maximum weight it can hold. Using its full capacity more often is one way to work within a tight profit margin.

The key to getting ultimate packout ratios is distributing customers associated with long travel times across multiple routes. The distribution enables workers to fill trucks quickly, dump, and get back on their routes—not easy to do on a continuous long haul as there isn’t time to packout trucks. But with well-planned, evenly distributed courses, haulers achieve packout ratios of 85-90% to 100%. Callen says that the higher percentages translate to less trekking to and from the landfill and more time knocking off collections.

 

Workday Utilization

Workday utilization is the percent of the day spent completing a route, divided by scheduled hours in the day.

With seven hours typically dedicated to the job, there is little slack to tug on to expand routes. As you aim to increase productivity, be careful to avoid long days and overtime, Callen advises, especially considering you must factor in weather, truck issues, and fluctuations in set out weights, among other often unexpected circumstances that add time.

Further, Krumski cautions, “You can only be behind the wheel so many hours a day, or you fall into a Department of Transportation safety violation.”

Ensuring evenly balanced workloads helps. Krumski looks at performance data to identify drivers who may finish in eight hours and those spending 10 hours on the road.

“When I see this disparity, I ask, where and how can we change up routes for a better workday balance and get people in simultaneously? For instance, if someone broke down, another driver can pick up the load.” He looks at automation, asking if he can change any part of the route to automatic side loaders (ASL) to rely less on pickers.

 

Service Time

Service time is hours spent only driving the route and collecting. That’s the most obvious job, but only part of what workers do during a collection day.

Haulers do their best to maximize service time. But mitigating factors weigh in, Callen says. Workers have about four hours a day to focus solely on collecting, spending their remaining time traveling to and from disposal/recovery facilities, waiting in line there, servicing their vehicles, on required breaks, etc.

One best practice is to shoot for route times that are 30 minutes shorter than the planned workday. Here, automation may again come into play. Asking customers to schedule bulk pickups saves time too.

 

Route Balance

Design routes to maximize weight, fuel, miles, and time.

Krumski leverages this KPI to explore if and how he might redistribute stops on each route to be as uniform as possible while considering these four factors.

An example is having two trucks serving the same route. Due to their size, the trucks have limited maneuverability, sometimes only able to pick up on one side of a street. When two trucks serve the same route, they don’t need to double back or drive around several times.

A route balance entails diving into multiple metrics. Krumski exemplifies this with a client scenario: there are two routes; one with 1,000 stops; another with just over 500. But they are balanced because there is less distance between the 1,000 stops.

“The route with fewer stops drives several miles uninterrupted, so the picker can’t ride on the back, which takes time. If you have several hundred consecutive close stops where the picker can ride on the step and quickly get into and out of the truck, you’re fine,” Krumski says.

So besides stops, he looks at the distance between stops, time to complete a route minus downtime, and especially watches whole-route weights.

“If weights are wonky and routes with heavier loads are trailing, that’s when we focus more on weights to balance routes,” Krumski says.

“But while weight is a big factor, it’s not everything, as seen in comparing the two routes where the one with the lesser units stopped much more frequently. And sometimes, weather or different human elements throw a wrench in your plan. I saw a lot of that during COVID.”

 

Collection Day Balance

This KPI refers to the range of time between the minimum and maximum cumulative times spent servicing all routes on a collection day. Getting that range entails looking at the fastest and slowest routes for each day.

A major discrepancy between the fastest and slowest crews calls for evaluation. Is it a routing issue, a collection issue, or a human issue? And is an adjustment needed?

Collection days and routes should be adjusted when one day requires an additional truck.

“Let’s say you have a tri axle servicing a route in two dumps. It goes down, so you have to send in two smaller trucks. Or you need a smaller truck to navigate alleys or side streets due to detours or other circumstances requiring negotiating smaller spaces. You use a larger truck for main routes and a smaller truck for problem areas,” Krumski says.

Distance and cost to dump also commonly come into play. Sending multiple trucks to dump once may save money over having one larger truck dump multiple times.

 

Keeping up with Changes

While KPIs quantify performance and help inform best practices, achieving good outcomes requires keeping up with changes. Ongoing training is a must.

“When I ask drivers why they made a given decision, they routinely say it’s what they were taught. That instruction sometimes comes from someone who hasn’t been with the company in years. Best practices have to evolve to keep up with changes in community development and new technologies,” Krumski says.

The trash industry excels when it evolves as a whole. Using KPIs and existing technology has great potential to influence change and improve daily routes.

 

You’ll find more collection route optimization information here:

 

 

 

 

 

 

 

 

 

 

 

 

Posted by Diane Samuels at 6:00 am