On-time in full (OTIF) is a key variable in beverage sales. Like most fast-moving consumer goods, beverage deliveries need 100% OTIF, day in and day out, to prevent losing revenue and market share to competitors whose private and dedicated fleets can better supply retail customers.
Beverage sales, therefore, are inextricably tied to fleet uptime and efficiency. To improve these critical measures, some of the most successful name-brand fleets are using pre-built predictive models to monitor a wide range of vehicle and engine failure types and make proactive repairs.
The pre-built models harness the power of data science and cloud computing to identify maintenance needs, days, and weeks, in advance of failure. In many instances, the insights from the models identify pending failures before fault codes appear. With these insights, fleets no longer must rely on fault codes or on static mileage and time-based intervals to inspect vehicles and schedule repairs.
Accurately diagnosing vehicle and engine failures at the earliest possible stage holds the key for beverage fleets to achieve greater uptime and availability of assets. With predictive insights, repairs can be scheduled and completed between shifts to keep vehicles operating in the normal rotation.
Gain the data advantage
Accuracy is the linchpin of effective predictive maintenance and workflow automation. Pre-built models that identify the root cause of a pending failure, with a very low margin of error, make it possible to standardize diagnosis and repair processes and improve shop efficiency as well as do jobs correctly, the first time.
By combining accurate insights with a standardized repair process and system integrations, fleets can quickly move beyond these all-too-common scenarios:
- A driver returns from a shift, enters the shop, and tells a technician about a mechanical issue. “I heard a knock in the engine and the vehicle does not have enough power.” The technician connects a scanner to the vehicle. Diagnosing the problem takes time and the results are often inconclusive.
- The maintenance department is overwhelmed by data from vehicle telematics systems that send hundreds, and perhaps thousands, of fault codes every day.
- Full-time employees are creating spreadsheets with pivot tables to organize fault codes and vehicle data by severity and location. The same employees are manually creating work orders and emailing their documents to technicians.
Create work order automation
Predictive models that integrate with your telematics and maintenance software systems will create a direct path from diagnosis to repairs. This will greatly improve the speed of repairs and efficient use of shop resources.
When using predictive models, private and dedicated fleets can maximize utilization of technicians and shop resources by following three practices:
- Create subject matter experts (SMEs). Designate a team to manage the insights and teach technicians to use the information to complete work orders correctly and efficiently.
- Reduce steps for repairs. Create a standard repair procedure for each predictive insight. For instance, you could document repair procedures for your Top 10 issues and create work orders immediately when insights are received by email or directly into your fleet maintenance system through an API.
- Maximize technician efficiency. To optimize work for technicians, some fleets have large monitors in shops that list inbound work. This gives technicians heads-up information to prepare for vehicle arrivals and expedite repairs.
Work order automation is possible with system integration between predictive models, fleet telematics, and maintenance management software. When technicians close out work orders in the maintenance program, the integration feeds the algorithms of predictive models with repair data to continually learn and improve the accuracy of results.
Get ahead of fault codes
Modern vehicles and engines have numerous failure points. In many cases, the signs of failure appear before fault codes. Aftertreatment systems are among the most serious failure types. Predictive models can identify the warning signs to prevent roadside breakdowns and expensive repairs.
Beverage delivery operations are prone to aftertreatment system failures. Engines may not be reaching high enough RPMs and operating temperatures to initiate passive or rolling “regens” to clean the diesel particulate filter (DPF). Knowing this, beverage fleets often have drivers or technicians do a manual regen while trucks are parked at terminals.
If manual regens are not cleaning the DPF completely, the vehicle could unexpectedly derate its power or shut down completely on the road. By monitoring engine data, a predictive model can identify problems, such as a stuck valve that is limiting the release of diesel exhaust fluid (DEF), before it’s too late.
Fleets can use predictive insights to address the root cause of aftertreatment system problems days and even weeks before failures. Other types of failures that pre-built models can identify include:
- Exhaust gas recirculation (EGR) valve
- DPF reductant heater
- EGR cooler
- Coolant contamination, coolant leaks, and thermostat failure
- Battery (low voltage), cabling, and alternator.
- Low oil pressure
- NOx sensor
- And much more
Calculate a fast ROI
The return on investment from a SaaS-based predictive modeling solution comes from the cost savings of minimizing downtime. By avoiding unplanned maintenance events and road calls, fleets will generally save more than $600 per vehicle per day. This does not include rental or towing fees, or lost revenue from missed deliveries.
With predictive insights, fleets can also save by completing more repairs in-house and avoiding the extra cost and downtime of dealerships. Additional ROI is created by increasing revenue from higher vehicle availability and by fuel savings from better vehicle performance.
The cloud-based Uptake Fleet predictive maintenance platform comes with pre-built models that simplify the process of proactively scheduling critical repairs. By analyzing historical work orders the models dynamically identify component-level “survival” curves. Also, by integrating with fleet telematics systems, the models detect anomalies from real-time data before fault codes appear.
Uptake Fleet insights have accuracy rates of more than 95%. Fleets that use insights from the platform to make better decisions are seeing double-digit reductions in maintenance costs and breakdowns, while increasing vehicle availability by similar amounts. Give Uptake Fleet a test drive.
