The Magic Behind Planning and Executing (Optimal) Service Supply Chains - Part 1

The recent three-part series entitled “Navigating Between Service Management Scylla & Charybdis” analyzed the phenomenon of service economy, or the increasing importance of the service sector in industrialized economies. But while the vast customer service software market’s opportunity was examined there, the series also pointed out the treacherous complexity of planning and executing service supply chains.

Compared to manufacturing and distribution supply chains, which are already complex on their own, service supply networks add a few more complicating variables. The series introduced the notion of the “perfect response” in service chains (vs. the “perfect order” in manufacturing and distribution) or how to get the right person with the right part and knowledge at exactly the right time to solve a particular customer’s need. And all that perfect response has to repeatedly and continually happen for almost every customer while maintaining a financially viable service business.

Every service business is ultimately measured by its performance on the day of service, and ensuring that the company has the right resources in the right places at the right times is critical. But the ability to deliver service on the day it is actually required is only the last link in a long chain of decisions that must be made weeks or months before the day of service.

Namely, manufacturing and distribution supply chains have several levels of planning (i.e., strategic, tactical, operational, and execution) with associated time horizons. These time brackets range from years (in the case of strategic business planning) to quarters, months, and weeks (in the case of tactical sales and operations planning [S&OP]master production scheduling [MPS], and operational materials requirements planning [MRP]) to hours, minutes, and seconds (in the case of real-time shop floor or warehouse control/manufacturing execution systems [MES]).

Distinguishing Between "Immediately Before" and "Well Before" the Day of Service

To run their business with some expectations (rather than on mere hope), service supply chains have similar planning levels and time horizons, which can be roughly divided into the following two time brackets:

  1. An immediate period around the day of service

  2. Forecasting and planning for the day of service

The immediate period around the day of service encompasses all of the necessary activities for the service company to be ready for slightly before, during, and after the day of service. First of all, there is appointment booking for non-emergency installations, repair, and preventive maintenance, whereby the time is locked, but technicians’ allocations might still be tentative. Then there is the execution at the day of service via scheduling tasks and hours/minutes within the day.

Last but not least, some scheduling and rescheduling of individual resources (and possibly time slots) often happens based on new emergency calls and unforeseen events in the field. The crux of the matter on the day of service is what time slot to schedule for each customer and which service engineer (or crew) to assign for providing the service. (The word “resource” will be used throughout this series to refer to any type of mobile field worker including engineers, technicians, inspectors, surveyors, gang, crew, shared vehicle and so on, while “job” or “task” refers to any work that these resources must perform.)

As for the long-term planning for the day of service, there are the following activities based on the planning time horizon:

  • Shifts within days – Staffing individual resources to shifts (rosters)

  • Days within weeks/months – Deciding on availability of individual resources for full days or weekly shifts (including sending staff to training, approving vacation requests, using contractors, using overtime, etc.)

  • Quantity for long periods – Deciding on the size and skill mix of the company’s workforce in the various territories for next quarter/year and discerning the forecast patterns of customer demand in each territory by type of service call.

The service chain is therefore a decision-making process consisting of several interrelated steps, which all ultimately affect the day of service and provide the information that must be available to service planners and dispatchers much earlier. Forecasting, planning, and rostering are all about thinking ahead, whereas scheduling is all about executing and delivering actual service slightly before and on the day of service.

While the results of scheduling decisions may be more apparent, the outcomes of planning decisions usually impact more jobs and resources. Planning decisions play a bigger part in the bottom-line performance of the service business (i.e., financial profitability or net income). In other words, by omitting or neglecting any of these important logical steps, any company’s ability to deliver customer service will likely be impaired.

Scheduling and Executing: Where the Rubber Meets the Road

It is likely that every service business experiences the operational service chain optimization challenge on a daily basis. Simply said, service chain optimization entails managing resources in the most efficient and cost-effective manner to deliver adequate levels of customer service that aid both customer satisfaction and retention. The “easier said than done” adage certainly applies here.

ClickSoftware Technologies’ recently published book “Service Chain Optimization for Dummies” points out that every scheduling decision has at least six major factors to consider before assigning a resource (and this consideration needs to occur simultaneously with all other outstanding jobs). In other words, before making a scheduling decision, a service organization should ask itself the following questions:

  1. Who? Which resources are available, with the necessary skills and qualifications, the required security clearance, and are they within a reasonable distance of the customer? Is more than one resource required?

  2. Does What? What job will the resource be doing? What skills are needed? How much time should the job take?

  3. With What? What parts, tools or special equipment are required?

  4. When? At what time can the resources arrive? What is the service level agreement (SLA) or appointment window? What is the customer’s availability?

  5. Where? Where is the location of the job? Is there a street address? Are there any geographic challenges to consider in the routing decision?

  6. For Whom? Who is the customer and do they have any special preferences or a special status?

Written succinctly, the service manager has to constantly decide "Who/does What/with What/When/Where/for Whom." Professor Moshe BenBassat (founder, current chairman, and CEO of ClickSoftware) named this principle “W-6” over two decades ago.

The challenge arises from the complexity of finding the best overall resource to tend to each job, every time, and every day. That is not necessarily just the closest resource, or the one who can necessarily respond the fastest, but the one who best balances the workload, minimizes costs, is skilled, and has the parts available to complete the job on the first attempt.

Scheduling gets even more difficult because one cannot schedule jobs in isolation: one rather needs to consider the next job, and the next one, and so on and so forth (throughout the entire day) to find the most efficient schedule possible. The individual decisions one makes in scheduling a resource directly and often irreversibly influence what follows.

Scheduling could be compared to a game of chess; once one player has committed a resource to a job, the future scheduling possibilities are altered so he/she must anticipate a few moves ahead to avoid vulnerability of his/her position and options. The optimized scheduling and mobile dispatch of service resources plays the most critical role in service chain optimization and produces the quickest and most apparent return on investment (ROI).

Approaching the Day of Service Challenge

Service businesses take many approaches in trying to overcome the service chain optimization challenge. Generally speaking, these approaches on the day of service could fall into the following three categories: compromise, automate, and optimize.

Many service businesses have traditionally made compromises to enable easier decision-making. In small enterprises with only a few field technicians, every technician is able to schedule himself/herself. In a somewhat more complex scenario of several small service teams consisting of up to five technicians, calls are channeled from the dispatch center to the so-called team buckets, whereby technicians share the calls within buckets and make impromptu decisions themselves.

But for environments with a few hundred technicians, to improve manual scheduling, a simple common compromise example is the creation of invisible boundaries of coverage, which are established solely to make manual scheduling more manageable. This means that any new job that arises within a territory is assigned to one of the resources that look after that particular territory. Technicians are organized into small territories (rigid boundaries) of, say, 50 resources per region, with the following rule: “Do not schedule calls from region A to technicians of region B!”

Establishing these invisible boundaries is a compromise that reduces the number of possible choices and the complexity of the decisions made by people (i.e., the dispatch team). Compromise is the lowest level of evolution for a service business trying to cope with the service chain optimization challenge. Spreadsheetswhiteboards, and simple scheduling applications are often used to support this working practice.

Yet, while compromising may appear to be simple and practical, it comes at a cost. This approach is inefficient because it disregards the overall workload balance, and does not allow scheduling of a closer and more appropriate resource from a neighboring territory. The all too common unfortunate result is one resource being over-utilized while a comparable resource in another territory is sitting idle.

The over-utilized resource could accumulate additional overtime costs whereas the underutilized resource is still being paid for sitting still. Ultimately, this approach might harm all aspects and metrics of customer service delivery: utilization, efficiency, productivity, and response times.

The more a company fragments its resources into territories and line of business (LOBs), the lower the overall utilization and productivity. This risks damaging the company’s level of customer satisfaction because it has less flexibility when delivering service.

Automation: a Stop-gap Measure

The next level of the service business evolution is to automate, whereby a simple computer program is used to automatically assign jobs to resources by addressing only some of the business’ considerations. For instance, the application will match skills requirements, but over-simplify other variables such as minimizing travel distances by using inaccurate aerial (as the crow flies) routes.

An example could be a service place with a few hundred technicians organized by LOB, e.g., installation technicians and repair technicians. The simplifying rule could then be “Do not schedule calls of type X to technicians of type Y!” Simple automation with limited intelligence can even be worse than manual decision-making itself because frequent illogical suggestions by the computer will make an experienced dispatch team question all of its other decisions.

This distrust in the system will cause the involved humans to manually intervene and eventually stop using the system altogether. Automation might bring some benefits with small improvements in utilization, efficiency, productivity, and response times. While it is an improvement over the abovementioned compromise approach, service chain optimization is still far from being achieved.

When people let computers make decisions on their behalf, they ordinarily expect these machines to come up with the right decisions. But to achieve this nirvana one needs optimization capabilities. This becomes especially apparent in service environments with thousands of resources, where in principle any kind of service call can be assigned to any technician as long as it makes “business sense.”

How Service Optimization Works

Optimization utilizes advanced and sophisticated scheduling algorithms to dissect the service chain problem and solve it in the most efficient and intelligent way. This approach is different from data processing applications that present the user with data and let the user make the decision.

Computer-based algorithms are powerful mathematical techniques working behind the scenes that are designed to solve complex problems quickly and efficiently. They are devised and written by mathematicians and almost always outperform humans.

Service chain optimization algorithms balance customer satisfaction and operating costs. Utilization, efficiency, productivity, and response times are all maximized while being consistently aligned with the company’s business strategy and objectives. Any conflicts between competing goals are resolved according to the company’s defined service policy (which will be covered in the next part of the series).

Optimization algorithms not only create the best starting point, but they also constantly review and revise the day’s plan to maintain the best possible schedules in an ever-changing environment. Scheduling is not just about placing and moving magnetic tiles on a whiteboard; it is rather about doing it most effectively and in the most compact manner to maximize utilization and optimize the customer experience.

Potential Benefits from Service Optimization

Optimization can create significant value and benefits for the service business in the following ways:

  • Reduced travel between assignments (i.e., by clustering jobs)

  • Minimal use of overtime

  • Mitigated risk to customer satisfaction through earlier arrival times

  • Increased resource capacity

  • Defined and predictable lunch breaks (to better manage both customers’ and field technicians’ expectations and satisfaction)

One of the usual business benefits that service businesses seek from service chain optimization is an increase in the productivity of their resources. For example, the company could increase the number of jobs successfully completed each day from four to five per resource, a modest 25 percent productivity gain. This is certainly one of the most transparent measurements of success in many, but not all, service businesses.

In an Aberdeen study from 2008, 8o percent of the surveyed organizations had the first-time-fix rate of 55 percent, i.e., 45 percent of their service calls required at least one more visit. However, the 20 percent of the organizations that Aberdeen labels as "best in class" claimed that only 14 percent of the calls are not resolved on the first visit. Imagine the advantage of these organizations over the rest of the industry--far better service and at far lower costs.

How do they achieve this three-times-better job completion rate? The answer is typically a combination of optimized time-and-route scheduling (so that the service engineer is not late for the appointment to find that the customer is not at home), optimized work scheduling (matching the engineer's skills and the engineer's in-van tools and parts to the job's requirements), optimized appointment booking, and real-time mobile data combined with real-time decision making (to handle the deviations from the plans, such as delays, absences, emergency jobs, etc., and instantly create new plans).

The next part of this series will illustrate how service excellence can be achieved on the day of service, starting with the setting up of a viable service policy. In the meantime, please send us your comments, opinions, etc. We would certainly be interested in your experiences with this software category (if you are an existing user) or in your general interest in  evaluating these solutions as prospective customers.
comments powered by Disqus