Street maps, particularly in developing communities, change frequently. Construction projects, traffic jams and vehicle accidents all conspire to wreak havoc on work schedules, while the CSR worries about maintaining efficiency, productivity and most importantly, customer satisfaction.
Before optimized scheduling and SLR, dispatchers made decisions about sending field resources to specific jobs largely based on dead reckoning. CSRs gave customers full- or half-day-long appointment windows for arrival because of the uncertainty about travel times as well as technicians' time on previous jobs. These same CSRs had to field the calls from angry customers when technicians were either late or did not show at all.
Street-level routing gives dispatchers and managers the information to make smart scheduling decisions. It removes the guesswork so routes can be mapped efficiently - ensuring technicians spend more time completing jobs and less time driving.
This article discusses the benefits of SLR, and how service organizations can use this approach to increase revenues and strengthen customer relationships. It also outlines new technologies that bring more intelligence to current street-level routing tools, so that organizations can make real-time decisions based on current traffic situations.
Getting a bead on directions
Location-based services such as street-level routing give service organizations "eyes" on how to best direct field technicians from job to job. Specifically, SLR blends detailed, current maps with GIS points. When combined with effective workforce management technology, SLR enables companies to boost productivity and minimize travel time by planning routes across the entire field workforce.SLR eliminates the inherent problems with linear distance or "as the crow flies" route planning. Congested neighborhoods, slow-moving highways, one-way streets and natural boundaries such as rivers or mountains all slow travel - despite the fact that two job sites may be close in linear distance. Figure 1 shows how linear-based routing sends drivers to jobs in close geographic proximity, but over slow-moving streets that extend travel time and drag down the schedule.
When service organizations incorrectly estimate travel time between two points, they pay more for fuel, vehicle wear and tear, periodic maintenance, etc. Worse, estimated arrival times change, ultimately affecting customer satisfaction. Suppose a mobile worker has six jobs to handle today and the company has accurately estimated travel times for all jobs except for between the third and fourth task.
If the company is lucky, it overestimated the travel time and merely lost the chance to increase productivity and possibly revenues by squeezing in yet one more task. Underestimation means the technician will likely be late to the last three jobs, and may not even make the sixth job that day. Because organizations set out to match field technician capacity to workload, it is unlikely a dispatcher will find an available worker close enough and not committed to another job to "jump in" and take one of the remaining tasks.
If a solution exists, it probably requires a reshuffling involving several employees, or the postponement of a lower priority task to the next day. These decisions, whether made by optimization software or by human dispatcher, face so many constraints that they can't yield a solution as effective as an optimized schedule based on accurate travel estimates.
The consequences of bumps in the schedule aren't good:
- losing customers
- failure to meet the terms of the SLA (service level agreement) - leading to decreased customer satisfaction, financial penalties and sometimes creating legal liability
- paying overtime as a result of bumping the schedule beyond standard working hours
- having to dispatch higher-cost workers such as subcontractors
- losing employee confidence in the scheduling process
- missing key performance indicator targets, which may lead to issues with regulatory governance organizations
Considering SLR alternatives
Some companies try to eliminate linear distance issues with alternatives to SLR. Scheduling software that incorporates maps will ignore street-based calculations of estimated travel times and distances (SLR), and simply provide a map with the locations of tasks and mobile workers. The software will provide directions and rough estimated drive times between two points on the route. But the software will not use those travel time estimates to make decisions about which technicians should go where.
Other approaches rely on postal code estimations: grouping tasks based on their proximity within and between postal codes. While this approach reduces the computational demand of SLR calculations, it is extremely inexact. The theory is that all points within a given postal code are considered to be close together, so the drive time is thought to be negligible. Two significant problems exist with postal code estimates. First, postal code areas can vary greatly in size. One point on the west side of an area may be five miles to the center (the reference point for calculating travel time and distance) and 15 miles to the furthest point on the east side of the same area. Second, travel time is calculated by dividing the distance by a single speed parameter, not taking into account congested neighborhoods, varying speed limits, etc.
Optimizing one route at a time using SLR cuts down computational demand, but also leads to incomplete information and inefficient scheduling. In this approach, the system calculates linear distances between tasks before using SLR to develop a more efficient route (based on proximity of tasks) for each worker. By not dynamically accounting for optimized routes across the entire workforce, the company can't capitalize on potential efficiencies presented by such factors as faster routes, proximity of workers with special skill sets, avoidance of natural boundaries and congested neighborhoods, etc.Pre-calculating travel between all service locations may be considered a shortcut to SLR provided the service locations are known in advance, the road map does not change much and the number of service locations is not excessive. The theory is to do all the calculations up front, and then refer to the stored results to optimize routes. That said, the calculation can be a problem. For example, if a company provides services to consumer products like refrigerators that may be in any of 100,000 homes in a suburban area, that company would have to calculate 10 billion different routes. One hundred computers figuring 20 distances per second would need a half year to complete the task.
These alternatives to street-level routing, along with the limitations of linear distance estimates, highlight SLR's real benefit to service organizations. SLR is not mainly about how a specific mobile worker should drive from point A to point B; its key productivity improvement is about deciding which mobile worker should be dispatched to point B.
Eye in the sky
Service optimization technology incorporating SLR can now leverage new data points such as real-time traffic updates to make sure field technicians avoid the traffic jams and other problems that delay on-time delivery and affect the rest of their schedules. This approach enables service organizations to dynamically select the shortest travel time - not necessarily the shortest distance - to minimize time on the road and maximize productivity and efficiency.
Most existing SLR systems rely on static GIS points, and can't predict if new construction or a vehicle accident will cause a 30-minute delay. By incorporating real-time traffic updates into their SLR-enabled workforce management systems, companies can change drivers' routes on the fly to avoid traffic delays caused by accidents, construction, traffic jams and other unexpected events.
In this approach, the company's workforce management system receives online traffic updates on an as-needed basis, and compares that information to the GIS routes between jobs. If the traffic updates indicate a problem, the scheduling system can automatically reroute field technicians to a faster route via dispatchers or direct messages to mobile devices. As with any workforce management solution, the technology should be flexible enough to let the company set rules on how to handle different scenarios. For example, managers can determine whether to have the system automatically reroute field technicians and push the schedule out (based on the anticipated delay), or do it manually. Larger organizations with busy field workforces will likely choose to automate the process, while smaller organizations may choose to handle traffic-related situations on a case-by-case scenario.
Like having a personal traffic helicopter watching every route, online traffic updates tied to SLR-enabled workforce management systems are another important step towards the creation of the real-time service enterprise, where mobile workforces are more efficient, productive and responsive to customer needs.
Creating a plan
Adopting some form of street-level routing for field workforce scheduling and management is no longer a luxury for most service organizations. It's an absolute necessity and a competitive advantage in markets where time and productivity mean everything. Organizations should consider true SLR that optimizes route scheduling across the entire workforce, so the system automatically assigns which technician, with what skills and tools, goes to which job and at what time. That flexibility ensures workers spend more time completing jobs, which increases profits and customer satisfaction.