Part 1 of this series established that service supply chains have many planning levels and time horizons, which can be roughly divided into the following categories: the immediate period around the day of service, and forecasting and planning for the day of service. My blog post then expanded on the various approaches to the challenges on the actual day of service.
Part 2 delved more deeply into the intricate execution issues on the day of service, starting with optimized scheduling, or the “W-6” optimization challenge: Who, does What, with What, When, Where, and for Whom? Mobile communication was established as the best means for dispatchers to communicate the schedule and job details to the resources, receive updates, notify customer cancellations, and continually optimize their schedule in response to all changes. Additionally, astute dispatchers use location data from global positioning system (GPS) tools to continually optimize schedules and divert the right resources to an emergency job.
Yet, the service chain is bigger than just optimized scheduling on the day of service. Even the best optimization algorithms can hit an early limit if the rest of the chain is ignored. If a service company’s workload is, say, 150 percent of its capacity for an extended period of time, then no level of optimization can overcome this reality.
It is natural for outsiders to assume that problems occurring during the day of service come from poor scheduling or from the service business responding too slowly to changes. But has the company planned the right level of capacity, and does it have enough resources working and available to deliver the right level of customer service? Poor mid- to longer-term planning decisions can impair the company’s daily performance, causing problems which could have been avoided through better preparation.
The short-term corrective options available to service managers are limited, costly, and sometimes painful. For example, the company can use overtime, subcontract its services (outsource), relocate resources, cancel training, or cancel vacations, but none of these alternatives are attractive. Hence, capacity-related decisions need to be made long before the day of service.
Looking Well Before the Day of Service
Therefore, companies must plan ahead to employ enough resources to meet their customers’ demands. The benefit to any service business from implementing improvements to each part of the chain individually is smaller than the combined benefit of having a well-geared holistic service chain.
All parts of the chain provide benefits, but most service businesses tend to start by focusing on their most immediate pain – the daily scheduling challenge. Once this short-term pain is removed, their focus tends to move onto the longer-term and strategic issues.
The total resource capacity available to a service manager on the day of service is the result of a sequence of decisions made long ago. The further ahead of the day of service that the company makes its decisions, the more choice and less costly options are available to its managers.
The service chain decision-making process consists of several activities. Working backwards through time before the day of service, one has the following:
Rostering (Shift Planning). The company might know how many resources it needs, but what shift patterns and working hours should it use to provide the optimal coverage to its customers? What is the best policy and process for building rosters for any shift, for weeks and months into the future? Accordingly, the shift patterns are calculated and the shifts are staffed several weeks, or even months, ahead of the day of service.
The idea is for service manager to know how many resources are working for what hours on which shift pattern. Shift patterns may not change frequently but the resources on each shift do, and managers have to ensure they have enough resources working on each shift to deliver the expected level of service to their customers.
Tactical resource planning. Before planning the shifts for given dates, service managers must ensure that their resources are available to work (taking into account planned vacations or training). At this time, they are making minor adjustments to cover any gaps; for example, adjusting the level of subcontractor usage, or approved overtime to accommodate recent updates in the expected workload.
Capacity planning. Long before tactical resource planning, based on a forecasted workload the company must determine the size and the mix of skills in each territory or service area to ensure it has enough resources and skills to meet the expected level of demand, service level agreements (SLAs), and plans. The company also needs to consider other options that its business may have that could affect service delivery such as the launch of a new product or service to customers. At this stage, decision-makers have sufficient time to make decisions that change their capacity, such as hiring, training, contracting, or even laying off resources if business is declining.
Demand forecasting. For all of the aforementioned activities to happen, the company must start at the beginning with its demand forecast answering the important questions about how busy the company will be in the future (next month, quarter, or years ahead), taking into account historical trends, seasonality, cycles, and new business initiatives. Without a forecast, everything else in the service chain decision-making process is a wild and uneducated guess.
Enterprises must have a reasonable prediction of future demand so that they can accurately plan their resources and address any gaps. Management should discern what marketing or business initiatives are likely to produce more (or less) demand.
Logically, in an integrated service supply chain, changes to the forecast will automatically adjust the capacity plan and shift assignments, while changes to the capacity plan will alter resource availability in the schedule. Since 100 percent accuracy is impossible to achieve, companies should instead recognize that the process of planning ahead is of greater value than any plan itself, because service is quite unpredictable (as described in Part 2).
Service chain decision-making is not restricted to mobile workforces. Organizations with stationed workforce such as call centers, retail stores, and hospitals can also benefit from entire service chain optimization. They all need to solve the following business problem: what is the best allocation of people to shifts in such a way that customer demand is adequately covered, work regulations are complied with, and reasonable employee desires are taken into consideration.
Distinguishing Between Planning and Scheduling
To recap, in scheduling, one is matching real jobs to specific times and resources on the day of service, whereby more jobs may appear as the day progresses. The aim is to work out how best to achieve the individual day’s service objectives.
With planning, one is matching resource levels against a forecast workload with varying degrees of certainty. Still, these long-range resource-level decisions will later affect scheduling at “the moment of truth.”
ClickSoftware Technologies’ recently published educational book Service Chain Optimization for Dummies presents an example that illustrates the difference. Say a service business employs 200 resources, who each complete 4 jobs per day, which results with the daily capacity of 800 jobs. But recent growth in this business has its average daily demand running up at 1,000 jobs.
The service manager has placed several requests to recruit more resources but these are refused. As a result, there is a shortfall of around 200 jobs per day. Even the best and most sophisticated optimized scheduling cannot fully resolve a gap of this magnitude. This insufficient capacity is a matter of resource planning and not resource scheduling.
To contrast scheduling and planning even more, in scheduling, the total service resource capacity is a given, and managers use this figure to deliver daily service to meet the demand. But if the demand is higher than the capacity then there is a problem that can be only partly resolved using overtime, contractors, or reluctantly pushing out some customer jobs to a later stage.
On the other hand, in planning, the company is seeking to establish its total resource capacity in terms of how many resources it will need, with what skills, located where, and over what time periods. In shift planning (a.k.a., rostering) one is planning the resources against defined shift patterns.
Staying on Top with Business Analytics
Most service businesses monitor their key performance indicators (KPIs) using a variety of metrics, e.g., the number of jobs completed per day, utilization, travel time per job, labor time per job, first-time completion rate, and re-visit rate. To make good operational use of these measurements, they should be analyzed by time, geography, business line, product, or any other way that is pertinent to the business. One can get the most benefit from analytics by giving each person the metrics that are relevant to them.
The company’s KPIs must measure the performance of its business in accordance with its established service policies. Otherwise, management cannot know if the business performance is effective and if their service policies are the right ones. The best operated service businesses keep on top of their service policies. Given that over time any company’s business strategy and priorities can change, companies need to update their service policies to reflect this.
Service managers should scrutinize the service performance reports carefully and look for patterns. For example, the average number of jobs completed per day per resource is interesting, but so are the maximum and minimum numbers. Another typical figure of interest is which resources complete the smallest number of jobs per day, but one should not necessarily jump to conclusions.
Maybe these resources work in a rural area with poor infrastructure where the travel distances between jobs are high, which reduces the number of jobs they can realistically complete each day. Maybe demand for their skills has dropped or maybe they are the company’s best resources, whom managers repeatedly assign to the toughest and longest jobs (and the company must look for ways to get others to learn from them). Drilling down and understanding relationships in this manner might uncover many ways of improving the company’s overall service performance.
Putting Service Metrics Into Continuous Improvement Perspective
Monitoring the performance of any service business during the day of service is very important so that managers can keep on top of the daily issues, react to any changes, and still meet their targets. Monitoring is a tactical use of performance data, but no less important is the company’s historic performance and how managers can use it in further optimizing the future performance of their service business.
Historic performance and demand data is the service supply chain’s feedback loop that helps to produce initial future forecasts. Needless to say, this future forecast is based on the assumption that history will repeat itself, and as we all know, this is not necessarily the case. Thus, the forecast must be adjusted to reflect any deviations from historic patterns and trends.
For example, consider the implications of average travel times rising in a service business. If the reported performance data is now the new standard, then managers must adjust their forecast to reflect this. Otherwise their total resource capacity for labor time is already unrealistic and overstated (because resources spend more time traveling and less time actually providing customer service).
Analytic review and answers to the questions such as “How did we do? Did we meet the goals of our service policy? Where can we improve? Who are our best and worst performers, and why?” are the end of service chain optimization, but they are also the beginning of the next service lifecycle.
Continuous improvement (CI) is about monitoring, predicting, and acting. Companies must ensure that they receive feedback on business performance to help refine future forecasts, plans, and service policies. Since it is never “one and done,” companies must use their service data and KPIs to further optimize their business processes.
Even if the enterprise believes it has a perfect service business model now, mid- to long-term changes in the market and changes in the business, and even short-term sharp changes in the weather can require an adjustment to the company’s service policy. Enterprises should continually keep looking for ways to improve delivering service to their customers. They should leverage the aforementioned smart technology for much more than just to automate today’s manual processes.
But to get the most out of their service chain optimization technology investment, companies must thoroughly rethink and change the way they operate internally. Dear readers, what are your comments, opinions, etc., in this regard? After two exhaustive series on service management, 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.