Process Analysis, Mapping and Business Process Management

Introduction

Operations are experienced by customers as processes: a request is received, information is checked, work is performed, a decision is made, and a product or service is delivered. Quality problems rarely arise from one isolated activity alone. They often emerge from the way activities are sequenced, handed over, resourced and controlled. Process analysis therefore provides the foundation for improving quality, cost, speed, flexibility and customer value.

This chapter examines practical methods for understanding and improving processes. It explains flowcharts, SIPOC diagrams, value-stream maps and swimlane models; introduces business process management (BPM); and develops key operating measures including cycle time, throughput, takt time, work-in-process and process efficiency. The chapter also shows how to identify bottlenecks, waste, variation and improvement opportunities.

A process should be analysed at an appropriate level. A high-level map can reveal ownership and interfaces, while a detailed map can expose rework, delays and control weaknesses. Effective analysis combines visual representation, quantitative evidence and the voice of the customer.

Learning Objectives

By the end of this chapter, students should be able to:

1. Define a process and distinguish processes from functions and projects.
2. Construct and interpret flowcharts, SIPOC diagrams, value-stream maps and swimlane models.
3. Explain the BPM lifecycle and the role of process governance.
4. Calculate cycle time, throughput, takt time, work-in-process and process efficiency.
5. Apply Little’s Law under appropriate steady-state conditions.
6. Identify bottlenecks, waste, handoff problems and sources of variation.
7. Recommend process improvements using evidence-based reasoning.
8. Evaluate ethical, technological and managerial issues in process redesign.

1. Understanding Processes

A process is a linked set of activities that transforms specified inputs into outputs for an internal or external customer. Inputs may include materials, information, labour, equipment, energy and decisions. Outputs may be physical products, services, records or outcomes.

A process has several essential characteristics:

• a defined trigger or starting condition;
• inputs and suppliers;
• a sequence of activities;
• resources and controls;
• outputs and customers;
• performance measures; and
• a responsible owner.

A functional department, such as purchasing or finance, is an organisational grouping. A process, such as “approve and pay a supplier invoice”, may cross several departments. This distinction matters because local optimisation can damage end-to-end performance. For example, a purchasing department may reduce purchase price by using a supplier with long lead times, while production experiences stockouts and expediting costs.

Processes may be classified as:

A process hierarchy helps select the right level of analysis. An enterprise value chain may contain a process family, which contains an end-to-end process, subprocesses, activities and tasks. A map that is too high-level hides causes; one that is too detailed becomes difficult to interpret and maintain.

Process owners and stakeholders

The process owner is accountable for end-to-end performance, even when activities are performed by different departments. Stakeholders include customers, employees, suppliers, regulators, owners and communities. Their requirements may conflict. A customer may seek speed, a regulator may require extensive verification, and employees may need safe workloads. Process design must therefore balance quality, compliance, efficiency and human factors.

2. Process Mapping Techniques

2.1 Flowcharts

A flowchart represents the sequence and logic of a process. Common symbols include:

A good flowchart uses action-oriented labels, such as “Verify customer address”, rather than vague labels such as “Address”. Decision branches should be labelled clearly, for example “Yes” and “No”. The chart should show rework loops, approvals, waiting points and exceptions rather than only the ideal path.

Flowcharts are useful for identifying unnecessary decisions, duplicated data entry, missing controls and loops. However, they may not show who performs each task or how long each task takes. Those limitations are addressed by swimlane models and value-stream maps.

2.2 SIPOC diagrams

SIPOC stands for Suppliers, Inputs, Process, Outputs and Customers. It is a high-level, usually five-to-seven-step view of a process. SIPOC is particularly useful at the beginning of an improvement project because it establishes scope without prematurely debating detailed procedures.

A SIPOC exercise should begin with the process output and customer requirements, then work backwards to inputs and suppliers. Inputs should be specific enough to test. “Information” is weak; “validated delivery address” is more useful. Customers include downstream internal users when their work depends on the output.

SIPOC is not a detailed operating procedure. It is a boundary-setting and stakeholder-alignment tool. It can reveal that different teams have different definitions of the same output, such as “completed order”.

2.3 Swimlane models

A swimlane diagram places process steps in lanes representing departments, roles, systems or organisations. The horizontal or vertical lanes make handoffs visible. For example, an expense-claim process may have lanes for employee, line manager, finance and payment system.

Swimlanes help answer three questions:

1. Who performs each activity?
2. Where does the process cross a boundary?
3. Where can accountability, information quality or waiting time be lost?

A handoff is a transfer of work, information or responsibility. Handoffs are not automatically wasteful, but each creates risk of delay, misunderstanding, duplication or failure. A strong swimlane model records the information or document transferred, not merely the department name.

2.4 Value-stream mapping

Value-stream mapping (VSM) examines the complete flow of material and information required to deliver value. It is associated with lean management and distinguishes value-adding time from waiting, transport, inspection, storage, rework and other forms of non-value-adding activity.

A current-state VSM commonly includes:

• customer demand and delivery requirement;
• supplier and customer information;
• process boxes;
• inventory or queue quantities;
• information flows;
• process time and changeover information;
• lead time or waiting time; and
• a timeline comparing total lead time with processing time.

A value-adding activity changes the product or service in a way the customer requires and is done correctly the first time. Necessary but non-value-adding activities, such as legally required inspection, may still be essential. The objective is not to remove controls indiscriminately; it is to reduce avoidable delay and improve the method of control.

The classic lean wastes are often remembered as defects, overproduction, waiting, unused talent, transportation, inventory, motion and overprocessing. Waste should be treated as a symptom to investigate. Excess inventory, for example, may be a response to unreliable suppliers or unstable schedules rather than the original cause.

3. Business Process Management

Business process management is a systematic approach to discovering, modelling, analysing, redesigning, implementing, monitoring and continually improving organisational processes. BPM connects process design with business strategy and governance.

A practical BPM lifecycle is:

1. Identify: determine important processes, customers, owners and strategic relevance.
2. Discover: document the current state using observation, interviews, records and process-mining data.
3. Model: represent activities, decisions, roles, systems and information flows.
4. Analyse: measure performance, variation, risk, cost, compliance and customer experience.
5. Redesign: develop and evaluate improved alternatives.
6. Implement: train people, configure systems, revise controls and manage change.
7. Monitor: track performance and exceptions using agreed measures.
8. Improve: learn from results and update the process.

BPM differs from a one-time improvement project because it establishes continuing ownership and measurement. It also differs from workflow automation. Automation may accelerate a poor process; BPM asks whether the process is necessary, correctly designed and governed before automating it.

Process governance

Process governance defines who may change a process, how risks are assessed, what documentation is controlled and which measures trigger action. A governance structure may include a process owner, subject-matter experts, quality or compliance representatives, information-technology specialists and frontline employees.

A process architecture provides a consistent map of major processes and their relationships. Process models should have version numbers, owners, effective dates and review intervals. Uncontrolled diagrams can create conflicting instructions and audit failures.

Process mining and digital BPM

Process mining uses event logs from enterprise systems to reconstruct actual process paths. It can reveal variants, loops, bottlenecks and compliance deviations that interviews may miss. Its results depend on the completeness and accuracy of event data. A system timestamp may indicate that a record was created, not that work actually began.

Artificial intelligence can support classification, demand forecasting, anomaly detection and document extraction. It should not be treated as an unquestionable process authority. Organisations must validate models, protect confidential data, explain material decisions and retain human accountability, especially in employment, credit, healthcare and safety-related processes.

4. Quantitative Analysis of Processes

Visual maps explain structure; measures explain performance. Measurements should use a clearly defined unit, population, time window and process boundary.

4.1 Cycle time

Cycle time is the elapsed time required to complete one unit, or the time between completion of consecutive units at a process output. In a stable repetitive process, these interpretations can be closely related, but analysts should state which definition is being used.

For an activity with processing time and waiting time:

where CT is cycle time, PT is processing time and WT is waiting or delay time. All variables must use the same time unit. If a process includes several sequential stages, total elapsed cycle time is the sum of each stage’s processing and waiting times, provided the stages and boundaries are defined consistently.

4.2 Throughput

Throughput is the rate at which conforming units are completed.

where TH is throughput, Q is the number of completed conforming units and T is the observation period. Units may be orders per hour, patients per day or claims per week. Throughput is not the same as capacity: capacity is the maximum feasible output under stated conditions, whereas throughput is the output actually achieved.

4.3 Takt time

Takt time represents the pace required to meet customer demand.

where A is available production time during the demand period and D is required customer demand during that period. Both must refer to the same period and exclude planned non-working time consistently. If a shift has 7.5 net working hours, or 27,000 seconds, and demand is 90 units, takt time is:

Takt is a demand-based planning rate, not a guarantee that every unit will take exactly that time. It should not be confused with cycle time. If cycle time is greater than takt time, the process cannot meet demand without additional capacity, reduced work content, improved method or changed demand assumptions.

4.4 Capacity and bottlenecks

For a single resource, a simplified capacity estimate is:

where A is available time and CT is time per unit. This assumes consistent availability, demand and quality, and ignores setups, failures, breaks and variability unless they are included in A or CT.

A bottleneck is the constraint that limits the effective throughput of the system. It is not necessarily the slowest individual activity in every period. A resource may be slow but have little demand, while another resource with a smaller cycle time may be overloaded because all work must pass through it. Bottleneck analysis should examine utilisation, queues, changeovers, downtime, yield, skill availability and demand mix.

4.5 Work-in-process and Little’s Law

Work-in-process (WIP) is the number of units that have entered a process but have not yet exited it. It includes units being worked on and units waiting, provided they are within the defined process boundary.

For a stable process over a sufficiently long observation period, Little’s Law states:

where WIP is average work-in-process, TH is average throughput and CT is average time in the process. Units must be compatible; for example, units per hour multiplied by hours gives units.

The relationship applies to a stable system in which average arrival and departure rates are effectively equal and the averages are measured over an appropriate period. It does not require a particular probability distribution, but it does require a consistent boundary and meaningful long-run averages.

Worked example. A claims process completes 24 claims per day and the average claim remains in the process for 2.5 days. Then:

If management reduces WIP to 48 claims while throughput remains 24 claims per day, the implied average cycle time is:

This is an operational implication, not proof that reducing WIP alone will reduce cycle time. If WIP is cut without addressing a bottleneck, throughput may fall or queues may move elsewhere.

4.6 Process efficiency

A useful time-based measure is process cycle efficiency (PCE):

where VA is value-adding processing time and LT is total lead time from entry to exit. The measure is meaningful only when value-adding criteria and process boundaries are explicitly defined.

Worked example. A service request spends 18 minutes in activities that directly create the requested service and 162 minutes waiting, checking, transferring or being queued. Total lead time is 180 minutes.

A low PCE does not mean that every non-value-adding activity can be eliminated. Some verification, safety and regulatory steps are necessary. The result signals an opportunity to examine delay and control design.

4.7 Quality-adjusted output

When defects and rework are significant, gross throughput can mislead. A process may complete many units but deliver few acceptable outputs. A simple first-pass yield (FPY) is:

For multiple independent stages, an illustrative rolled throughput yield is the product of stage yields:

This multiplication assumes that each yield is defined consistently and that the stage relationship is appropriate. For example, three stages with yields 0.98, 0.95 and 0.97 have:

Thus approximately 90.3% of units pass all three stages without the defined failure condition.

5. Finding Improvement Opportunities

Waste and failure demand

Improvement begins by separating symptoms from causes. Common signals include queues, repeated data entry, approvals with no decision value, excessive movement, frequent expediting, rework, customer complaints and unused employee capability.

Failure demand is work created because something was not done correctly or completely the first time. Examples include customers calling to correct an invoice, staff reopening incomplete cases and technicians returning because parts were missing. Measuring failure demand helps reveal the true cost of poor quality.

Root-cause analysis

Useful methods include the five whys, cause-and-effect diagrams, Pareto analysis, stratification and direct observation. The five whys should not be used mechanically; the investigation should test each proposed cause with evidence. A Pareto chart can prioritise categories, but a large category is not automatically the easiest or most valuable cause to address.

Designing countermeasures

Potential countermeasures include:

• simplifying forms and decisions;
• eliminating duplicate entry;
• moving controls closer to the source of error;
• balancing work across resources;
• reducing batch size;
• standardising work while allowing justified exceptions;
• improving layout and information visibility;
• using pull signals and appropriate WIP limits;
• automating repetitive, rule-based tasks; and
• redesigning roles and interfaces.

The preferred solution should address the constraint and protect quality. A local improvement that merely shifts work or creates hidden risk is not an end-to-end improvement.

6. Illustrative Case: Reducing Delays in an Equipment-Repair Process

This is a fictional illustrative case designed for learning purposes.

Background

A university facilities department repairs laboratory equipment. Users submit requests through an online portal. The process includes request logging, triage, approval for external parts, technician diagnosis, repair, testing and closure.

Problem

Users report that repairs take an average of 8 working days. Technicians state that the repair itself usually requires only 2.5 hours. Management initially proposes employing another technician, but the process team decides to map the complete flow before investing.

Analysis

A SIPOC exercise identifies laboratory users as customers, the portal and technicians as major inputs and suppliers, and a tested, returned instrument as the required output. A swimlane model shows that requests move between the portal administrator, laboratory manager, stores, technician and finance office. The largest handoff occurs when technicians wait for approval of parts.

A sample of 40 cases shows:

Little’s Law provides a consistency check. With throughput of 5 cases per day and lead time of 8 days:

This agrees with the observed average WIP. The time-based process cycle efficiency, using an eight-hour working day, is approximately:

The low value indicates that waiting dominates elapsed time; it does not imply that all checking is waste.

The current-state map identifies three causes: incomplete request information, a twice-weekly approval batch and poor visibility of parts availability. Rework is also generated when the instrument is returned without a recorded test result.

Approach and solution

The team introduces a standard request form with mandatory equipment identification and fault description. Low-value, predefined parts are placed under a delegated approval limit. Parts status is made visible in the work-management system. Requests are triaged daily rather than twice weekly, and a standard closure checklist requires a test result and customer notification.

A pilot is run for four weeks. The team monitors lead time, throughput, WIP, first-pass closure and customer complaints. It does not judge success by lead time alone because faster closure with more repeat failures would reduce quality.

Outcome

In the pilot, average lead time falls to 4.5 days, average WIP to 23 cases and first-pass closure rises to 94%. Throughput remains approximately 5 cases per day, indicating that the main improvement came from reducing waiting and rework rather than adding capacity. The department retains approval controls for high-risk and high-cost parts.

Learning points

1. End-to-end mapping can challenge an assumed capacity problem.
2. Little’s Law links observed WIP, throughput and lead time.
3. Handoffs and approval batching can dominate elapsed time.
4. Quality measures must accompany speed measures.
5. Delegation should be risk-based, documented and auditable.

7. Implementation, Ethics and Limitations

Process maps are models, not reality. They simplify complex work and may omit informal knowledge, exceptional cases or emotional aspects of service. Analysts should validate maps with the people who perform the work and compare claims with records and observation.

Measurement can create dysfunctional behaviour. If employees are rewarded only for speed, they may omit checks, avoid complex cases or manipulate timestamps. Measures should therefore include quality, safety, customer outcomes, employee workload and compliance where relevant.

Process redesign may change jobs, authority and access to information. Ethical implementation requires consultation, fair treatment, appropriate training and transparent communication. Personal data used in process analysis should be limited to legitimate purposes, protected from unnecessary disclosure and retained according to applicable policy and law. Automated decisions should be reviewable, and affected individuals should have a route for correction or appeal where appropriate.

A process must not be optimised beyond its control boundaries. Removing an approval may reduce time but increase fraud exposure. Eliminating inspection may increase defects. Reducing WIP may make a system fragile if demand is highly variable. Improvement decisions should consider risk, resilience and the total cost of ownership.

8. Best-Practice Checklist

Before approving a process improvement, ask:

• Is the customer and required output clearly defined?
• Is the process boundary explicit?
• Has the current state been validated with process participants?
• Are actual data and exceptions included?
• Are cycle time, throughput, WIP and quality measures defined consistently?
• Has the bottleneck been distinguished from a visible symptom?
• Does the proposed change improve the end-to-end process rather than one department?
• Are safety, legal, privacy and ethical controls protected?
• Are ownership, training, documentation and monitoring arrangements clear?
• Will the process be reviewed after implementation?

Chapter Summary

Process analysis makes the flow of work, information and responsibility visible. Flowcharts describe sequence and decisions; SIPOC diagrams define high-level suppliers, inputs, process, outputs and customers; swimlane models reveal roles and handoffs; and value-stream maps connect material and information flow with lead time and waste. BPM provides a lifecycle for governing and continuously improving processes.

Cycle time measures elapsed completion time, throughput measures completed output per unit of time, and takt time represents the pace required by customer demand. WIP, throughput and cycle time are related by Little’s Law under stable, consistently defined conditions. Process cycle efficiency compares value-adding time with total lead time, while yield measures help ensure that speed is not achieved at the expense of quality.

Effective improvement combines mapping, direct observation, reliable data, root-cause analysis and stakeholder involvement. Bottlenecks, queues, rework and handoffs should be analysed end to end. Technology and AI can strengthen BPM, but they require sound data, governance, human accountability and protection of privacy and safety.

Key Terms

Process: A linked set of activities that transforms inputs into outputs for a customer.

Process owner: The person accountable for end-to-end process performance.

Flowchart: A diagram showing process sequence, decisions and flow direction.

SIPOC: A high-level view of Suppliers, Inputs, Process, Outputs and Customers.

Swimlane model: A process map that assigns activities to roles, departments or systems.

Value-stream map: A representation of material and information flow, including processing and waiting time.

Business process management: The disciplined lifecycle of discovering, analysing, redesigning, implementing and monitoring processes.

Handoff: A transfer of work, information or responsibility between parties.

Cycle time: The elapsed time required to complete a unit or the interval between completed units, as defined by the measurement boundary.

Throughput: The rate of conforming output completed by a process.

Takt time: Available time divided by required customer demand.

Work-in-process: Units inside a process that have not yet exited.

Bottleneck: The constraint limiting effective system throughput.

Little’s Law: The steady-state relationship WIP = TH × CT.

Value-adding activity: An activity that changes the output in a way the customer requires and is done correctly the first time.

Process cycle efficiency: Value-adding time divided by total lead time, expressed as a percentage.

First-pass yield: The proportion of units passing without rework or repair.

Process mining: Analysis of event-log data to reconstruct and evaluate actual process behaviour.

Failure demand: Work generated because an earlier output was incomplete, incorrect or unsuccessful.

Process governance: Rules, roles and controls for owning, changing and monitoring processes.

Review Questions

Short Answer Questions

1. What is the difference between a process and a functional department?
2. What information does a SIPOC diagram provide?
3. Define takt time and state how it differs from cycle time.
4. State Little’s Law and identify its variables.
5. Why are handoffs important in process analysis?

Long Answer Questions

1. Explain how flowcharts, SIPOC diagrams, swimlane models and value-stream maps complement one another.
2. Describe the BPM lifecycle and explain the importance of process governance.
3. Discuss the meaning, assumptions and managerial use of cycle time, throughput and capacity.
4. Explain how bottlenecks and waste can be identified using both qualitative and quantitative evidence.
5. Evaluate the ethical and operational risks of automating a poorly designed process.

Analytical / Application Questions

1. A service process completes 36 cases per day, and each case remains in the process for 1.5 days. Calculate average WIP using Little’s Law. State the assumptions required.
2. A production line has 25,200 available seconds per day and demand of 84 units per day. Calculate takt time. If the constrained workstation requires 330 seconds per unit, assess whether it can meet demand without improvement.
3. Construct a SIPOC and a swimlane model for a university student-enrolment process. Identify two likely bottlenecks, two forms of waste and three measures that should be monitored.

Multiple Choice Questions with Answers

Question 1

Which tool is primarily used to define Suppliers, Inputs, Process, Outputs and Customers at a high level?

A. SIPOC diagram

B. Control chart

C. Pareto chart

D. Gantt chart

Correct Answer: A. SIPOC diagram

Question 2

A process completes 30 units per hour and has an average cycle time of 2 hours. Under steady-state conditions, what is the average WIP?

A. 15 units

B. 28 units

C. 60 units

D. 90 units

Correct Answer: C. 60 units

Question 3

Available time is 21,600 seconds per day and customer demand is 72 units per day. What is takt time?

A. 180 seconds per unit

B. 300 seconds per unit

C. 360 seconds per unit

D. 432 seconds per unit

Correct Answer: B. 300 seconds per unit

Question 4

Which map most directly highlights transfers between departments or roles?

A. SIPOC diagram

B. Swimlane model

C. Inventory record

D. Pareto chart

Correct Answer: B. Swimlane model

Question 5

A process has 12 minutes of value-adding time and 120 minutes of total lead time. What is its process cycle efficiency?

A. 1%

B. 10%

C. 12%

D. 90%

Correct Answer: B. 10%

Question 6

Which statement best describes a bottleneck?

A. The activity with the highest purchase price

B. The constraint that limits effective system throughput

C. Any activity performed by more than one employee

D. The first activity in a process

Correct Answer: B. The constraint that limits effective system throughput

Question 7

A process has stage yields of 0.98 and 0.95. Assuming the defined conditions for multiplication apply, what is the rolled throughput yield?

A. 0.9305

B. 0.9600

C. 1.0300

D. 1.9300

Correct Answer: A. 0.9305

Question 8

What is the main purpose of a current-state value-stream map?

A. To display employee salaries

B. To show material and information flow, including waiting and processing

C. To replace all operating procedures

D. To calculate annual profit only

Correct Answer: B. To show material and information flow, including waiting and processing

Question 9

Which measure counts units that have entered a process but have not yet exited it?

A. Takt time

B. Throughput

C. Work-in-process

D. First-pass yield

Correct Answer: C. Work-in-process

Question 10

Why should quality measures accompany speed measures in process improvement?

A. Faster work always increases defects

B. Speed alone can encourage omission of controls or creation of rework

C. Quality measures are required only in manufacturing

D. Speed cannot be measured directly

Correct Answer: B. Speed alone can encourage omission of controls or creation of rework