Process analysis for Operations Management
Jan Riezebos, November 2003
Introduction
Operations management is responsible for fulfilling customer demands. It designs, plans, operates, controls and improves the day-to-day transformation processes within an organization in order to fulfill these demands. Operations management decides on the arrangement and utilization over time of engineering, purchasing, production, inspection, and transportation processes, facilities, and employees involved in the transformation. In some organizations this includes the order intake and billing process as well.
The operations function of an organization is designed to achieve a specific performance level. This is in general a mixture of goals, for example: 95% of the orders for product A have to be delivered within 2 weeks after order issue, 98% of the orders for product B have to be delivered within 1 week after order issue, no more than 1% of the orders of both product types may exceed the due date with 4 days or more, operational expenses per week should be lower than 50.000, customer complaints with respect to the quality of the products delivered may concern no more than one order a month, etcetera. The performance requirements are not always explicitly stated, but the actual design of the operations function resembles a specific mixture of the perceived performance objectives.
The operations function operates in a dynamic system and environment. An organization has to respond to or anticipate on changes in technology, products, and markets by redesigning their operations function or reformulating the performance objectives. It requires an analysis of both the operations function and the required performance mixture to determine the specific modifications that have to be made.
The field of business administration provides several tools for analyzing the fit between required performance and the system. Many of these tools lack a clear focus on the operations function of the organization. Instead, they focus on modifications in issues such as organizational structure, culture, product innovation, and the type of technology applied. Often, these modifications cannot be realized in the short term. However, there are also some basic instruments that help to analyze the operational function of an organization and provide insight in the type of modifications needed when a gap is detected between required and actual performance of the system. In operations management, we apply these instruments within the context of a process analysis instead of a functional analysis.
Processes are the main objects of study at the operations management level. A process transforms input into output. It consists of a network of activities (operations) that may depend on each other. It is initiated by an external signal (e.g. the receipt of an order), and it is performed on a more or less frequent basis. Within a process one or more actors are involved. A process has limited resources available (including time) and a specific mixture of objectives. Activities in a process are both transformations in shape, location, or condition, and in time and money. They can be examined from the
perspective of their contribution to the objectives of the whole process. The performance of a process is at least partially measurable, which makes it possible to control the process.
Essential elements of this definition of a process (actors, activities, dependencies, resources, objectives) are also found in the definitions provided by Davenport & Short (1990: 12) “logically related tasks performed to achieve a defined business outcome”, and Harrington (1991:9) “any activity or group of activities that takes an input, adds value to it, and provides an output to an internal or external customer”. However, we also include the initiation and control of the process in a process analysis. Figure 1 shows the basic building blocks in a process analysis.
Transformation process Planning and Control
Mixture of objectives
Performance measurement
Information flow Goods flow Initiation
Figure 1 Elements in a process analysis
According to Wijngaard (2000), a process analysis consists of four stages, as shown in Figure 2.
1. Objectives and performance 2. Identification relevant processes
3. Within-process gap analysis per performance dimension 4. Analysis extra-operational causes for performance gaps
Figure 2 Stages in a process analysis
The first stage is oriented towards objectives and performance. It consists of two separate activities, (a) determine the mixture (relative importance) of objectives for the process, and (b) measure the actual performance. It results in determining whether there is a gap between the actual performance and the mixture of objectives.
This is the basis for the next step in the process analysis. With the specific performance gaps in mind a global description of the transformation process is provided as well as a description of the control of the process. This description is focused on identifying the relevant sub processes that have to be taken into account in the detailed analysis. For these relevant processes, the main phases, activities, actors and resources are denoted.
The third step provides a detailed analysis of the causes for the gaps that have been detected in each performance dimension. This detailed analysis is limited to a within- process analysis. It starts with a description for each performance dimension. The analysis examines both the configuration of the process and the way it is operated and controlled.
Finally, the fourth step relates the shortcomings that were found in the current process design to causes outside the process. These causes are denoted as extra- operational.
The next sections will give attention to these four stages of a process analysis in operations management.
Stage 1: Objectives and performance
Schönsleben (2000: 12) distinguishes five performance dimensions that are relevant for business processes: Quality, Delivery, Finances/Costs, Flexibility, and Information/Knowledge/Know-How. The latter dimension is not an explicit performance objective in an operations management view of a production process, so other authors (e.g. Slack et al. (1998) who distinguishes Quality, Speed, Dependability, Costs and Flexibility) ignore this objective.
The meaning of these performance dimensions varies per situation. Based on insight in the characteristics of market and production situation the prevalent meaning of the performance dimensions can be stated more exactly. For example, flexibility in an engineer-to-order situation will not refer to the number of different products that can be delivered, as this is not a useful definition of range in such a production situation. Range can be defined in terms of different technologies that can be used or different raw materials that can be processed when transforming materials into finished goods. Tools that are available for this step in the analysis are e.g. the product/process matrix of Hayes
& Wheelwright (1984), product life cycle analysis, order qualifying and order winning criteria, and market analysis. Slack et al. (2001: 45-56) gives excellent examples of the different meaning of a performance dimension in different situations. It requires a clear understanding of the business to formulate these more exact interpretations of the performance dimensions, but is a prerequisite for a thorough analysis. It should result in a description of the measures per indicator, as shown in Figure 3 (Schönsleben 2000: 94).
Reason for measuring
Facts to measure Reference to business
object(s) Context specific definition
Flexibility Cost
Delivery Quality
Indicator
Reason for measuring
Facts to measure Reference to business
object(s) Context specific definition
Flexibility Cost
Delivery Quality
Indicator
Figure 3 Performance objectives explicated
Each process has to achieve a mixture of these potentially conflicting objectives.
Processes that aim at a high cost/quality ratio may perform less on dimensions such as delivery and flexibility. For processes that aim at short delivery, inefficient usage of resources may be unavoidable, resulting in higher costs. Therefore, the performance objectives for processes are interrelated. We can think of a process as a balloon with four forces simultaneously pressing on it, as shown in Figure 4. If one of the forces becomes stronger, the balloon transfers this pressure to the other three sides. The shape of the balloon changes, and hence process design need to be modified. Thus, a process is explicitly designed in order to achieve a mixture of objectives with different weights.
Quality
Delivery
Cost
Flexibility
Figure 4 Interrelatedness of objectives
One of the problems faced in practice is that the weights for the objectives are not explicitly stated. However, the actual requirements for a process do change over time, which will make it necessary to reconsider the design of the process. The next step in stage 1 is therefore to determine the main processes that are affected by these changes and their specific objectives. These objectives depend on the strategic objectives of the company, the markets that are served, and the environment. Changes in these factors will have impact on one or more processes. But due to the fact that many changes may have
occurred unnoticed since the design of a process, we cannot limit this search to recent changes in one performance dimension. All relevant performance dimensions have to be examined and per dimension one or more objectives should be determined.
Finally, the weights of the objectives have to be specified in a type of balanced scorecard. Schönsleben (2000: 94) gives attention to the specification of these objectives and to the problem of weighting these objectives. He explicitly relates the objectives to business objects, such as customer orders, work centers, etcetera. This level of detail is needed in order to continue the analysis and measure a possible gap between objective and actual performance.
The next step in stage 1 is to determine whether there is a significant gap between the objectives and the actual performance. Schönsleben (2000: 74) suggests performing this analysis at product family level, as each product family may have different objectives, i.e. lead time differences, quality differences, etcetera. Products within the same product family share to a large extend the same processes. If historic data is available, one can relatively easy apply a quantitative analysis. But sometimes it is already clear from a quick qualitative analysis (e.g. the list of customer complaints) that the system is not able to perform as required by a specific objective. Although this is sufficient to initiate the second stage of a process analysis, it helps to complete the analysis for all relevant objectives, as a redesign of a process will affect them all. This information will help to select the type of process redesign.
Stage 2: Identification relevant processes
The next stage in a process analysis in operations management results in focusing on the (sub) processes and control mechanisms that are primarily responsible for the performance gaps that have been detected in stage 1.
The analysis starts with a broad picture of the whole value chain, starting from customer order acquisition and finishing at delivery, installation, after-sales service, or billing. For this chain, a network of organizational functions that support and control this chain is depicted. The final picture includes all internal and external sub processes that have to be considered for their contribution to the performance objectives. Schönsleben (2000:105) introduces the organization-oriented process diagram for this purpose, but other representations (a goods flow diagram, a project network and organizational break down structure, etcetera) may be useful as well. Figure 5 gives an example.
Assembly Bidding
Wood parts Metal
parts
Delivery Detailed
engineering Assembly Billing
Bidding
Wood parts Metal
parts Wood
parts Metal
parts
Delivery Detailed
engineering Billing
Planning Inventory management Purchasing Planning Inventory
management Purchasing
Logistics Prototyping Inspection Logistics Prototyping Inspection
Figure 5 A broad picture of the value chain and supporting functions
Next, all components of this picture are examined whether they contribute to the performance gaps detected. The actual impact of a sub process or organizational function on each of the performance objectives is described in a way that helps to identify the necessity of including this sub process in the detailed analysis of stages 3 and 4.
In general, we expect the 20-80% rule to be valid, indicating that only a small portion of all sub processes will have to be considered in detail, as they contribute for the largest part to the main performance gaps.
The description of the actual impact of a sub process or organizational function on the performance objectives requires a basic understanding of the complexity of operating, planning, and controlling the whole process. This basic understanding can be obtained by representing the situation in a morphological scheme, a part of which is shown in Figure 6.
Production employees
A Number <20 20-40 41-100 101-300 301-750 >750
B Educational level
primary school general secondary
education
primary vocational
training
secondary vocational training
higher vocational
training
university
C Training days a year 0 1 2 3 4 >4
D Flexibility of employees low moderate high
E Task integration operating logistics maintenance quality transport
F Regulation people in charge of machines machine in charge of people
Production employees
A Number <20 20-40 41-100 101-300 301-750 >750
B Educational level
primary school general secondary
education
primary vocational
training
secondary vocational training
higher vocational
training
university
C Training days a year 0 1 2 3 4 >4
D Flexibility of employees low moderate
Production employees
A Number <20 20-40 41-100 101-300 301-750 >750
B Educational level
primary school general secondary
education
primary vocational
training
secondary vocational training
higher vocational
training
university
C Training days a year 0 1 2 3 4 >4
D Flexibility of employees low moderate high
E Task integration operating logistics maintenance quality transport
F Regulation people in charge of machines machine in charge of people
Figure 6 Part of a morphological typology of production situation
Schönsleben (2000:109-127) presents a morphological scheme that includes a characterization of product, production, planning, and order-related issues, including the location of the customer order decoupling point. In actual situations, additional categories that are relevant for the situation examined can be included to enrich this scheme. Using such a scheme, we can better formulate why a specific sub process or organizational function has such a high impact on a performance objective.
The selected sub processes should now be related to each other in a new picture and the causal relationships with respect to the performance gaps have to be indicated. The
new picture is used in the next stage of the analysis in order to depict the process of performance decay.
Stage 3: Within-process gap analysis per performance dimension
The third stage of the analysis aims at a deeper understanding of the reasons for each performance gap that has been detected. It takes both the internal structure of the sub processes and the relations between the sub processes in the final picture of stage 2 into account. Further, it will provide an indication of the type of process redesign needed.
The principal type of analysis in this stage is a detailed process analysis per performance objective. A popular metaphor for such a detailed process analysis is to
‘staple yourself to the order’. The business object that is present through all process steps is the (customer) order. By stapling yourself to such an order in the analysis, a detailed description can be given of :
y the activities performed (what, when), y the actors involved (who),
y the resources used (where),
y the signals that initiated the activity (why), and
y the units informed of the results of the activity (signal for next activities).
A process chart tool can be very useful in this type of analysis. Schönsleben (2000:
109) shows an example. The Actor-Activity Diagram (see Figure 7) that is studied in the second year of the Business Administration Bachelor program is a technique that also has proven to be very useful. For a description, see http://www.bdk.rug.nl/medewerkers/d.j.schaap/aad/index_oud.html. A geographical process diagram (flow depicted in physical lay-out, Schönsleben 2000: 108) will help to examine the process infrastructure.
Figure 7 Actor-Activity Diagram (source: website Schaap)
The process description for a specific performance dimension should start at the point where the promise for the outcome is made. Sometimes, negotiations with the customer on the specifications of this performance dimension are part of this process.
Promises on lead time, warranties, quality, and price can all be made in a negotiation process. These promises are input to the actual process of performance creation. The description should make visible how the process tries to achieve the promised performance. It will give attention to the sub processes that are involved in translating the promised performance into detailed specifications, process plans, resource allocation, and outsourcing decisions. Next it will give attention to the way the transformation is being realized. Special attention should be given to the issue of transfer between different actors, and losses that may occur due to this transfer.
Note that for each performance objective, a separate process analysis has to be performed, requiring different process descriptions. For example, an explanation for a too long lead time may have to be found within the process, resulting in a process description that pays attention to the time used for each activity (production, inspection, transportation, planning, issuing), as well as waiting times for the order and the main reasons for these waiting times. The actors involved in the activities and responsible for the waiting times have to be denoted. The same order may also face a quality gap.
However, the process description will be different. For a quality problem (e.g., number of customer complaints too high), we have to show the contribution of each process step (activity) to the final complaints. More attention will be given to the number of errors made in each step, the actual timing of these errors, the resources used, the transfer points, and the control mechanisms in place. Less attention will be given to the time
needed for the inspections. But for all dimensions, attention will be given to actors, activities, resources, dependencies, and disturbances and the way they are coped with.
The performance dimension flexibility requires special attention in a detailed process analysis. Flexibility can often not be measured with respect to one (customer) order, as it is not the result of a negotiation with a single customer. The process that results in a specific amount of flexibility is not launched by an order. Stapling yourself to one order is therefore not the correct way to identify the reason for a performance gap with respect to flexibility. It is a performance measure that relates to the whole product family (range) that has to be served by this process.
Flexibility can be decomposed into three components: range, mobility and uniformity (Upton, 1998: 17-19). The last two components specify range flexibility in terms of delivery (transaction speed), and quality (transaction loss). If we have to examine the range of a process (the set of different products that can be served by this process), we can analyse the reasons for not being able to extend this set. Some resources will probably not be able to manage a broader range than specified, due to technological limitations or lack of skills. By extra investments (which leads to a cost increase), the flexibility performance may be extended.
If we have to examine the mobility dimension of the flexibility of the process (the speed of changing between different products), we can model the changeover process itself, with speed as outcome. The process is launched by a request for a changeover, and we can attach ourselves to this request and describe why it takes so long before the changeover has been finished. By repeating this for several products, we do have information on the mobility dimension of the flexibility objective.
If we have to examine the uniformity dimension of the flexibility of the process (the yield or quality loss if transferring to different products), we can model the changeover process with quality as outcome. The reasons for a quality decrease or increase can be detected and assigned to the changes made during the changeover within the process infrastructure (resources), inputs (material), or control methods (inspection procedures). Note that the range dimension is affected by tightening either the constraint with respect to mobility or uniformity.
A detailed process analysis gives for each performance dimension insight in the possibility of formalizing the customer demand with respect to this performance dimension in order specifications, and the possibility of forecasting process behaviour. It can be accompanied with a quantitative analysis for a set of customer orders, as this results in more reliable (statistical) information on the behaviour of the process. In general, the analysis leads to the discovery of redundancies or incompatibilities within the processes and the planning and control applied to them (either in the way performed or in their design).
Literature on lean manufacturing provides a framework for describing various types of inefficiencies or redundancies found in processes. The Toyota production system distinguishes seven sources of waste (Suzaki 1987: 12-18):
1. Waste of producing defective parts and products 2. Waste of moving materials over long distances
3. Waste of building inventories
4. Waste of excess production (more than the market demands)
5. Waste of people waiting for materials, equipment, or other employees 6. Waste of inefficient processing of material (poor techniques, technologies) 7. Waste of unnecessary walking and hand movements by operators
These sources for waste are in the lean manufacturing literature mainly related to the performance dimensions quality, delivery, and costs. Flexibility as an objective translates in this view to delivery and quality objectives, i.e. changeover time. Literature on agile production gives more attention to design for flexibility.
Harrington (1991:134-159) translates these seven sources of waste to eleven tools that can be applied to general business processes. He focuses on improving throughput time, reducing disturbances, and reducing efforts to operate the processes as required.
1 Bureaucracy elimination Examine waste due to unnecessary administrative tasks, approval procedures, and paperwork.
2 Duplication elimination Examine waste due to identical activities at different locations in the process sequence
3 Value added assessment Examine waste due to performing activities that do not add value to the required performance
4 Simplification Reduce the complexity of the process 5 Process cycle-time
reduction Reduce the throughput time of the process
6 Error proofing Reduce possibilities to make mistakes in process steps 7 Upgrading Improve the usage of tools
8 Simple language Reduce the complexity of instructions 9 Standardization Reduce the variety in work procedures 10 Supplier partnerships Improve process inputs by supplier relations 11 Automation and/or
mechanization Replace routine tasks by automation or mechanization Figure 8 Process improvement tools (Harrington 1991:134-159)
Al these improvement tools are focused on partial improvements of a process. The inherent danger of using these tools is that the ‘improvements’ lead to undesired effects on the other performance dimensions. We should therefore not directly start with improving the process after the analysis for a single performance objective has been finished. First, the analysis of all performance objectives has to be completed.
Next, a check on consistency and cohesiveness of the improvement possibilities has to be performed. In case the suggested improvements all point in the same direction, implementation can start quickly. For example, some slack in one objective (e.g. an actual delivery performance of 99% in stead of the required 95%) may be used to fill a gap in another objective. The improvement suggestions for the two objectives point in the same direction. However, we will often discover improvement suggestions in one area
that will have negative impact on other dimensions. Such suggestions may be neglected if sufficient alternative improvements are available.
Finally, a decision has to be made what type of process improvement is needed. We distinguish both radical and continuous process redesign strategies. Radical improvement is focused on implementing a totally different way of operating and controlling the process, while continuous improvement is focused on improving the current process design. Using the metaphor of the balloon, continuous improvement reshapes the balloon such that it fits within the boundaries of the performance objectives, while radical improvement replaces the balloon for a new type with different characteristics and behavior.
Stage 4: Analysis extra-operational causes for performance gaps
Before a choice is made on how to redesign the process, we have to understand why the process was designed that way. The main reasons for the old design can often be found outside the operational process itself. Changes may have occurred in the information systems available for planning, controlling, and maintaining the process, educational level of employees, organizational structure, communication systems, etcetera. If these changes have a negative impact on the performance of the processes, we denote them as extra-operational causes for the performance gaps.
For example, if the type of labor work in the process does not fit anymore with the expectations of the current employees, it would not help if we simply replace one employee that makes too much mistakes for another. The process needs a more fundamental redesign in order to fit with the capabilities of the current employees. The within-process cause for the performance gap (an ill-functioning employee) relates to an extra-operational cause for this performance gap.
Engineering / Technology
Accounting / Finance
Human resources / Organization
Information technology / Communication
Marketing Product development
Operational causes
Figure 9 Extra-operational functional areas (modified from Slack et al. 2001: 30) Fields that can be investigated for detecting extra-operational causes for performance gaps are depicted in Figure 9 (see Slack et al. 2001: 30). Note that sub processes such as sales and billing are not denoted as extra-operational, as they belong to
the operational process. This holds also true for engineering activities that are performed in response to customer orders. However, availability of new technology and communication systems, new product development, maintenance, and marketing promotions will be considered extra-operational, as they are not involved in the realization process of a specific (customer) order.
Conclusion
This paper has presented a diagnostic scan for processes from an operations management point of view. The diagnostic tool has a clear focus on the operations function of an organization. Processes are the main objects of study in this diagnostic scan. It gives attention to the transformation processes (warehousing, manufacturing, delivery), administrative processes, such as order preparation, purchasing, engineering, and billing, as well as the planning and control of these processes.
This paper follows Wijngaard (2000) in distinguishing four stages for a diagnostic process scan. It starts with determining if there is a gap between the objectives and the actual performance. Generally there are a number of conflicting performance dimensions.
For this analysis, the performance dimensions have to be explicated such that they can be measured.
The next stage aims at detecting the processes that contribute to the performance gaps found. It starts with a broad picture of functions involved in the realization of performance, and characterizes the sub processes and its supporting functions in terms of complexity and contribution to the performance gaps detected. This results in selecting a small portion of sub processes that contribute for the largest part to the performance gaps.
The scan continues in the third stage with a detailed process analysis for each performance dimension by describing the selected sub processes in terms of actors, activities, dependencies (signals), resources, and contribution to that performance dimension. This analysis will result in discovering several causes for each performance gap. It will also provide insight in redundancies or incompatibilities within the processes and their control. However, these improvements are not directly implemented, as the diagnostic scan has not yet been finished. First, the causes that are found for each performance dimension have to be checked on consistency and coherence. And finally, a decision has to be made what type of process improvement is needed.
The fourth stage of the diagnostic scan aims at providing a better foundation for this decision on the type of process improvement by analyzing possible causes for the performance gaps outside the operational processes. We denote them as extra- operational. If such causes are detected, either a more radical change of the processes is required or a change in these extra-operational factors is needed. The diagnostic scan provides a sound foundation for this decision.
References
Davenport, T.H., Short, J.E., 1990, The new industrial engineering: Information
technology and business process redesign, Sloan Management Review 31(4), 11–
27.
Harrington, H.J., 1991, Business Process Improvement: The breakthrough strategy for total quality, productivity, and competitiveness. New York: McGraw-Hill.
Hayes, R.H., Wheelwright, S.C., 1984, Restoring our competitive edge. John Wiley.
Schaap, D.J., http://www.bdk.rug.nl/medewerkers/d.j.schaap/aad/index_oud.html.
Schönsleben, P., 2000, Integral logistics management: planning and control of comprehensive business processes. Boca Raton: St. Lucie press.
Slack, N., Chamnbers, S., Johnston, R., 2001, Operations Management, third edition.
Prentice Hall.
Suzaki, K., 1987, The new manufacturing challenge: techniques for improvement. New York: The free press.
Upton, D.M., 1998, Designing, managing, and improving operations. Prentice Hall.
Wijngaard, J., 2000, De PSM-scan, Faculty of Management and Organization, University of Groningen.
1 The Groningen Service Scan
Manda Broekhuis, Martin Land
The Groningen Scans provide frameworks for diagnosing respectively service and manufacturing processes. Typical for the Groningen Scans is the focus on (logistic) performance. The scans start by determining an appropriate objective in terms of performance. This enables a focused diagnosis of the business processes. The other typical element is the operations perspective. The scans are confined to identifying operational opportunities for improvement, which may relate to the primary processes, coordination, planning and control and their organisational embedding. In this Service Scan the determination of performance objectives is preceded by a step to determine the relevant service elements and customers/stakeholders. This step is straightforward for most manufacturing processes, but it may be complex and crucial in the case of diagnosing a service.
The Service Scan consists of four main steps (to be discussed in this document)
1) Determine the relevant service elements and relate them to customers/other stakeholders
2) Determine the performance aspects to be improved 3) Describe the processes
4) Analyse operational opportunities for improvement The previous steps will normally be followed by
5) Analyse non-operational opportunities for improvement 6) Redesign of (partial) processes
7) Determine requirements regarding management of change
Each step will be clarified for the example of the repair department of a large car retailer.
The tools to be used in each step are discussed in (Slack et al. 2001), (Metter et al.
2003), or in material provided during lectures.
2 1) Determine the relevant service elements and customers/other stakeholders
Explanation:
The performance aspects to be investigated in step 2 of this scan may depend strongly on what is identified as the service(s) to be investigated and who is being recognized as the customer of this process. Step 1 must provide suitable system boundaries in terms of service (elements) to be researched.
Sometimes multiple services may have to be included in the scan, sometimes - based on a project proposal or specific objectives - we may focus our project on one service process. Still a service can be affected by other services provided, when these make use of the same resources and/or are offered to the same customers. In turn, each service can be seen as a package with multiple elements.
Particularly in professional services a lot of other stakeholders may be identified, not typically being end-customers, but also imposing certain needs. Think of the service being provided by a physician in a hospital. The patient being treated is the main customer, but a young physician being educated will also be a stakeholder.
Tools:
- service package (slides, provided on paper) - circle of Grönroos (slides)
- stakeholder analysis Output:
- overview of relevant services and their elements - overview customers and other stakeholders Example:
The repair department of a car retailer delivers multiple explicit services such as the repair of broken cars (for an external customer, regular and non-regular), the maintenance of cars (for external customers/regular), the preparation of new cars for sale (for the sales department). Suppose we will focus on the repair of broken cars.
Then different markets with different requirements can still be identified. Performance requirements to be identified in the next step may differ among markets. In this respect, the repair of a car involved in an accident must be distinguished from the repair of a flat tire. In the former service, insurance companies may be customers/stakeholders as well, an external paint shop may be involved, etc. We only mentioned the explicit service of repairing a broken car. It is important to identify the other elements of the service package before performance indicators can be derived. This relates to (1) facilitating goods, such as car components; (2) supporting facilities, such as repair equipment and the waiting room; (3) implicit services, such as a nice atmosphere in the waiting room.
Each element of this package may have its own performance indicators.
3 2) Determine the performance aspects to be improved
Sub-steps:
a) Determine performance indicators for the relevant elements of the service.
b) Determine their relative importance c) Determine actual performance gaps
d) Determine the performance aspects to be analysed.
Explanation:
Step 1 resulted in a specification of the service elements to be researched (the object of research). Step 2 must lead to a suitable objective for the research. A good objective will specify a planned contribution to performance. In some cases the scan will be triggered by a perceived problem of the service organisation. Then, step 2 of the scan must relate the perceived problem to a ‘functional problem’ in terms of performance losses. In other cases the researcher must determine which performance objective(s) may contribute most to the goals of the organisation.
a) The first sub-step to be performed is the translation of general performance dimensions into operational indicators for each of the service elements specified in step 1. A suitable set of general performance dimensions to be used as a starting point in operations management are 1) costs, 2) speed, 3) dependability, 4) quality, 5) flexibility (see Slack et al.). The translation of these dimensions into operational indicators strongly depends on the service package being investigated. Each element of the service package will have its own indicators. For instance, speed may be indicated by the fill rate of facilitating goods, by the average waiting times of customers, by the time to solve a complaint etc. Particularly quality (4) is a complex dimension within services. In this scan we will focus on the aspect of accuracy (error-free delivery or conformance to standards)1. Flexibility (5) may relate to market heterogeneity and to demand fluctuations. The market heterogeneity indicates the extent to which different customers require different services and different levels of other performance indicators. Demand fluctuations may relate to changes of product, mix and volume in the course of time.
Notice that indicators of flexibility have a time and a range component: how quick can the organisation adapt to certain requirements, and to which extent?
b) Not every dimension will be equally important. It depends on the market and more specifically on the strategy being followed. Each market may have different ‘order winners’ and ‘qualifiers’. Thus, the relative importance of performance can be derived
1 Typical for quality (4) in services is a customer perspective, with internally for instance the Servqual dimensions reliability, responsiveness, assurance, empathy, and tangibles. In this perspective our dimensions speed and dependability will be perceived as elements of quality, respectively relating to responsiveness and reliability. Nevertheless, the perceived reliability of a service relates to both dependability and accuracy. Accuracy (error-free delivery or conformance to standards) is an element of service quality within the scope of operations management.
Generally, operations management actions have less influence on the other Servqual dimensions assurance, empathy, and tangibles, which find their roots in fields such as communication and human resource management. The management of customer expectations is also not within the scope; it will relate more to service marketing.
4 from strategies and market analyses. It can also be determined from interviews or surveys among managers or customers.
c) When it is possible, realised performance should be compared with desired performance for the relevant indicators in order to determine performance gaps.
Sometimes the effort to record actual performance or to get appropriate measures for desired performance may result in postponing this step and including it in the analysis.
To be able to determine the results of redesign, data on performance should at least be recorded before changes have been implemented.
d) The relative importance of performance dimensions combined with (b) the size of the performance gaps (c) may help in the choice of a performance objective. After a suitable performance objective has been determined, the research design can be specified.
Output:
- Operational indicators for performance dimensions
- Indication of relative importance of performance dimensions - Comparison of desired and realised performance
- Research objective in terms of contribution to performance improvement - Research design
Tools:
- Priority zones (p. 608 Slack et al.)
- Service Process Matrix (p. 9/10 Metters et al.)
- Criteria for evaluating the Service Package (provided on paper) - Servqual and the Gap Model (chapter 8 Metters et al.)
- Quality Function deployment (p. 175-178 Slack et al.) Example:
Step 1 included the decision to focus on the process of repair of broken cars. More specifically we will focus on the repair of cars (without external operations) that are offered for repair without appointment.
In sub-step (a) we must determine how to translate each performance dimension into operational indicators. We will confine ourselves to the dimension of speed. Most generally, the speed of the repair process relates to
- the time between offering the car for repair and completion of the repair.
When full repair is going to take a long time, another indicator may be important for speed:
- the ability to provide quick solutions to get on the road again.
From a customer perspective it will also be important to have - timely information about delays.
Looking at other elements of the service package will be important to guarantee sufficient
- availability of car components.
We will confine ourselves to speed. Notice that for other performance dimensions the supporting facilities, and implicit services may require their own indicators.
5 For sub-step b we interview the manager. He distinguishes two types of customers:
(1) customers who make use of multiple services: buying cars, having maintenance and repair at this garage.
(2) local customers who mainly come for car repairs.
For the first group ‘speed’ is the most important aspect of the service. For local customers ‘’ costs’ are more important.
For sub-step c we learn from the manager that the cost level is well-controlled. Most important is to improve the delivery speed for the first group of customers. Generally, the customer is promised to have his car back within one day (the same day, if offered in the early morning), but at least in 20% of the cases this is not realised.
For sub-step d this leads to the researching objective of decreasing the repair time for the distinguished category of repairs. Considering this objective a research design is developed to drive the following steps.
6 3) Describe the service process (such that the origin of the performance can be seen)
Sub-steps:
a) describe the characteristics of service requirements b) describe the characteristics of resource availability c) describe the relevant service delivery process
d) describe the processes of coordination, planning and control Explanation:
The description of the service should be such that it shows how performance emerges.
Therefore descriptions will strongly depend on the performance objectives defined in the previous step. Independent of the performance objective, important building blocks of an operational description are the elements depicted in the figure below.
Service delivery process
Co-ordination, planning, and control
Resource availability
Service demand Service
delivery process
Co-ordination, planning, and control
Resource availability
Service demand Figure : descriptive building blocks
First the characteristics of service demand should be investigated. The description of this first building block indicates the origin of complexity in realising performance. It should clearly identify the characteristics of volume, variety, variation and the degree of customer contact (see Slack et al.). The distinction between predictable and unpredictable parts of service demand, and controllable and uncontrollable parts will be important for most performance aspects. Market homogeneity and demand fluctuations are particularly important for the analysis of flexibility.
Whereas service demand indicates the origin of complexity in realising performance, resource availability determines the potential for realising this performance. Resources to be distinguished are material, personnel, information and supportive resources.
Flexibility, reliability and cost aspects of resources are important for the potential to realise a certain performance. Typical for services is that also the customer himself may be a relevant resource. Decoupling stocks are an important element of resources when responsiveness and efficiency (low cost) are being pursued. Notice that some information can be stored instead of being collected upon customer request. Also distribution of resources across FO and BO are relevant for the performance realised. To show the potential for flexibility, it is relevant to know whether resources are dedicated or generally applicable.
The service delivery process brings available resources and service demand together. It is important to identify those sub processes that affect the performance dimensions researched. Critical for most performance dimensions are the interfaces between processes: interfaces between customer and FO (the service encounters) but also
7 interfaces between FO and BO. A good service blue print will typically point out these critical moments in the delivery process. Also in service operations processes it is important to define the order decoupling point.
The processes of co-ordination, planning and control must contribute to having the right resources (and/or customers) in the right place on the right time. In the description it must be clear which resources need to be coordinated and what kind of coordination measures are used. The structures of these processes can be made explicit as a decision framework. Important for the quality of description is that - beside the procedure - inputs and outputs of each decision are clear and can be related to performance.
Output:
- blue print
- description of demand characteristics and resource characteristics - an initial idea of the causes of not attaining a desirable performance - overview of service encounters
- overview of relevant co-ordination, planning and control instruments and measures Tools:
- Service blue print (slides)
- Actor Activity Diagramming (AAD) (other courses) - Flow charting techniques (app 3, Slack et al.)
- Customer-processing framework (p. 172-173 Slack et al.)
- House of quality and Quality function deployment (p. 175-178 Slack et al.) - Design factor schemes (chapter 5 Metters et al., provided on paper) - All kinds of classifying schemes for service processes (Metters et al.) Example:
The description must focus elements that are relevant for the speed of car repair. Within the description of demand it is important to describe demand for other services as well.
These services make use of the same resources. Particularly maintenance provides a controllable part of demand. Certain patterns in the number of cars offered can be distinguished, providing some predictability.
The description of resource availability focuses on the availability of facilities and mechanics, the distribution of skills across mechanics and the stocking policies for car components.
In describing the delivery process, it is important to focus on these sub processes that are important to realise a certain performance. For our focus on speed, we must trace the relevant sub processes or activities, which may cause delays, such as the delivery of car components, a lack of information concerning customer requirements. This garage has problems with meeting delivery dates, so it makes sense to focus on the order accepting process, the planning, and the administrative processing of order information.
Flow charts and blue prints may point out critical activities for these processes.
8 The description of the planning process should describe the policies for planning appointments, the criteria for promising a certain delivery time, the feedback process in case of delay etc.
9 4) Analyse operational opportunities for improvement
Explanation:
The description was already directed towards showing the origin of performance. The analysis explicitly relates performance (gaps) to the elements of the description. Thus, it points out the causes of gaps or opportunities for improvements. If performance has not yet been quantified in step 2, this must become an element of the analysis. The type of analysis will depend strongly on the performance indicator to be investigated.
An analysis of cost performance should clarify the trade-offs between the current cost components, e.g. cost of overage versus cost of ‘walked customers’. Costs must be related to (the availability of) resources, including cost of customer efforts, and the efficiency of the distinguished processes in using resources.
The relevant indicators of speed (responsiveness) in step 1 should exhibit in which parts of the service delivery process responsiveness is most important. A first analysis of related response/throughput times could identify their composition. Generally, waiting times will be the largest component. But waiting times can be attributed to different causes. An in-depth analysis could distinguish the causes of the waiting times and relate them to the described characteristics of demand, resource availability, delivery process and co-ordinating and planning processes. Another analysis of responsiveness could identify to which extent decoupling possibilities are utilised and whether appropriate customer/FO/BO interfaces have been created. Notice further that responsiveness may also relate to service recovery.
Dependability partly relates to speed, but more specifically to control of throughput times. An analysis of dependability may show strong resemblance to the analysis of responsiveness, though more emphasis will be on processes of planning and control.
Besides, realised delivery times should be compared with promises made to customers It has already been mentioned that quality is a complex dimension within services. From a customer oriented perspective, speed and dependability will be perceived as elements of quality. The perceived reliability of a service incorporates both dependability and accuracy. An analysis of accuracy should first point out where and to what extent inaccuracies occur in each service element, and next trace their origin. Both resources, delivery processes and co-ordination may affect accuracy. For other elements of service quality, it should first be examined whether operations management can affect them.
Performance regarding assurance and empathy may have its roots in service operations, but often non-operational analyses (step 5) may help tracing their origin in for example the fields of communication or human resource management.
The flexibility analysis must relate the flexibility required by service demand to the flexibility provided by resource availability, delivery processes and co-ordination processes. Thus, it should point out where opportunities for flexibility have been left unused. More specifically, one may investigate whether predictable and controllable parts of service demand are sufficiently exploited within planning and control of the service process. In step 2 the internal dimensions of flexibility have been discussed. The distinction between time and range of flexibility can help in structuring the analyses.
Besides it is important to view flexibility at different time scales. That is, demand
10 fluctuations and variety within a day should be analysed separately from seasonal fluctuations and variety. An analysis whether the capacity of the customer himself is used sufficiently may be important for flexibility as this capacity is always available when it is needed.
Output:
- Overview of causes for performance gaps or opportunities for performance improvement
- Relative impact of causes Tools:
(Costs)
- Cost tradeoffs (chapter 9/10 Metters et al.) - Flow charting (app. 3, Slack et al.)
- Blue printing - AAD
(Speed/Dependability)
- Capacity strategy classification (chapter 2 Metters et al.) - Capacity policies/tactics classification (slides)
- Waiting time formula’s (chapter 12 Metters et al.) - Queue classification schemes (slides)
- Maister’s rules for managing perception of waiting time (slides) - Work measurement techniques (app. 2 Slack et al.)
- Recording techniques (app. 3 Slack et al.) (Quality)
- Gap model (chapter 17 Slack et al.)
- Cause&effect diagram (Fishbone/Ishikawa) diagram (p. 703-710 Slack et al.) - Poka-yoke (chapter 6 Metters et al)
- Conformance to specification, House of quality (chapter 18 Slack et al.) - Pareto diagrams (chapter 18 Slack et al.)
(Flexibility)
- Demand forecasting techniques (app 1 Slack et al.) - Pareto analysis (chapter 18 Slack et al.)
(General)
- (all kinds of) service classification schemes (Metters et al.)
Example
Our analyses focus on the speed in handling repairs. A first analysis determines the composition of the throughput times of repairs. A sample of repairs is followed in detail.
In a flow chart or blue print (created in the previous step) we show the average times of elements and the variances.
11 For each waiting time the causes are indicated. The large waiting time before the repair starts can be attributed to congestion. At 8 a.m. a large number of cars is delivered on appointment, but not every car can be handled instantaneously. The delivery of cars at the beginning of the day prevents the garage from idle times. By the end of the day the cars will be collected by their owners. A bottleneck analysis shows that facilities are hardly ever restrictive, most of the time the number of mechanics is the restricting capacity.
During the day, waiting times occur because of congestion, sometimes because the right components are not available. A number of times, the mechanic is called away for some small urgent repairs. Next, net repair times and waiting time can be compared. Generally the net repair time is less than one hour for a car ready after 4 hours. Most small jobs have been finished before the lunch brake. Typical cars with net repair times of about 2 hours appear to exceed the promised delivery time of one day. This asks for a further analysis of priority setting.
Another analysis determines the number of (unplanned) repairs per day. A comparison shows that this number is for more than 95% of days not higher than 50% of average capacity. Most of the capacity is filled with planned maintenance. Particularly a large number of cars is offered for repair on Monday mornings.
An important aspect of responsiveness is also that the car owners are informed timely when they cannot collect a car at the promised time. Interviews with the mechanics may show that they can generally determine after one hour whether the promised delivery time can be realised. However, it appears that most customers are informed in the course of the afternoon, after the floor manager has made his round. A further analysis should be made on the possibilities to improve the information to the customer.
Steps 5-7
Steps 5-7 of a service improvement project stretch beyond the scope of this document.
The reader is referred to tools provided by other functional management fields for analysis of non-operational causes of performance gaps (step 5) and the management of change (step 7). Ideas for redesign (step 6) regarding operational causes depend strongly on the results of step 4. As such, all kinds of (service) operations management concepts and methods can be relevant. Nevertheless general guidelines can be given to start the redesign. If a large number of causes for performance gaps or opportunities for performance improvement is identified in step 4, an impact analysis is necessary to determine their relative importance in (re)design. The relative importance of each should be weighted against the efforts/costs to improve a certain factor, when determining the focus of redesign. Besides, factors can be grouped based on commonality of underlying causes. For example, when a large part of causes/opportunities relates to an ineffective planning process, this may determine the focus of redesign. Besides clear boundaries and design criteria have to be specified before the design process is started.
12 References:
Metters R, King-Metters K, Pullman M, (2003); Succesful service operations management, South Western (Thomson), Ohio.
Slack N, Chambers S, Johnston R, (2001). Operations management (3rd ed.), Prentice Hall (Pearson Education), Harlow (England).
QUESTIONNARE
Questions 1 to 4 are designated to obtain information regarding the importance of performance objectives and elements of process in analyzing processes within the operations function.
1. What is your opinion regarding performance objectives as a starting point in analyzing processes within the operations function?
………
………
2. How do the scans (The Groningen Manufacturing Scan and The Groningen Service Scan) accommodate the performance objectives? In your opinion to what extent do the scans accommodate the performance objectives?
………
………
3. There are several elements of the process (i.e. activities, inputs, transformation process, outputs, resources, customers, interrelated sub-processes, owner of the process, and boundary of the process) that can be identified. Please write down the importance (priority number) of those elements to be accommodated within the scans according to your opinion!
……Activities
……Inputs
……Transformation process
……Outputs
……Resources (human and non-human resources)
……Customers
……Interrelated sub-processes
……Owner of the process
……Boundary of the process
4. How do the scans accommodate those elements of process in your opinion?
………
………
Questions 5 to 9 are designated to obtain information regarding the application of the
‘old’ PSM-Scan and the manufacturing scan in service organization based on users’
experiences.
5. Based on your experiences, which difficulties did you find when the ‘old’ PSM-Scan was used in a service organization?
………
………
6. Are there certain types of service organizations where the ‘old’ PSM-Scan does not fit to?
………
………
7. Which elements distinguish the good application of the scan from the bad?
………
………
8. Do you think the manufacturing scan can be applied directly to analyze service operations?
a. Yes b. No
And the reason is …...
9. If the manufacturing scan cannot be applied directly to analyze service operations, what modification should be made to the scan?
………
………
Questions 10 to 20 are designated to obtain information regarding the characteristics and the requirement that the service scan should has. Also these questions are designated to obtain information regarding the criteria needed as a base to improve the service scan.
10. In the service scan, unlike the manufacturing scan, the first step is the determination of the relevant service element (particular service process) and the customers, and then followed by the determination of the performance objectives. What is your opinion about the order of these steps?
………
………
11. How do you think the order of the steps should be?
………
………
