2. Production Data, Processes and Performance Architecture

Introduction and context

Production planning and control begins with a deceptively simple question: what must be made, how must it be made, and how long and how much will it require? Reliable answers depend on the quality of production data. A production system cannot plan materials accurately if its bill of materials is incomplete, cannot schedule operations if its routing is wrong, and cannot evaluate performance if work-centre capacities or time standards are unreliable.

Production data therefore forms the architecture through which customer demand is translated into materials, operations, resources, schedules, costs and performance results. The architecture links product definition, process design and managerial measurement. It is relevant in job shops, batch production, repetitive assembly, process industries and digitally integrated factories, although the form and level of detail vary across these environments.

This chapter examines the principal data structures used in production planning and control: process maps, bills of materials, routings, work-centre records and time standards. It then develops the performance measures needed to interpret operational results, including productivity, utilization, efficiency, throughput, lead time, service and cost. The central managerial issue is not simply to calculate these measures, but to understand their relationships and avoid optimizing one measure at the expense of the total system.

Learning objectives

After studying this chapter, the reader should be able to:

1. Explain how production data supports planning, scheduling, costing and control.

2. Construct and interpret process maps and bills of materials.

3. Develop routings, work-centre records and credible time standards.

4. Distinguish productivity, utilization, efficiency, throughput, lead time, service and cost measures.

5. Diagnose the operational causes of poor performance using integrated metrics.

6. Apply data governance and digital tools to improve production information.

7. Evaluate trade-offs among capacity, responsiveness, quality, inventory and cost.

The architecture of production information

A production system has three connected views. The product view describes what the customer receives; the process view explains how it is produced; and the resource view identifies the people, equipment, space, materials and time required. A planning and control system is effective when these views are consistent.

For example, a product record may identify a finished pump, its engineering revision and its approved specifications. The bill of materials decomposes the pump into a casing, impeller, shaft, seals, fasteners and purchased components. The routing identifies operations such as machining, balancing, assembly, testing and packing. Work-centre records describe the machines and labour groups capable of performing these operations. Time standards estimate setup, processing, inspection, movement and other relevant durations. Together, these records support material requirements planning, capacity planning, dispatching, costing and delivery promises.

The data must also possess integrity. A missing component may generate a material shortage; an obsolete routing may schedule work on an unavailable machine; an underestimated setup time may make a feasible schedule appear achievable when it is not. Data quality is therefore an operational capability rather than a clerical concern.

From product requirement to process map

A process map represents the sequence and relationship of activities required to transform inputs into outputs. It may be a high-level map showing order entry, procurement, production, inspection and dispatch, or a detailed map showing every operation, decision, queue and movement within a work centre.

A useful process map identifies:

• inputs, outputs and responsible parties;
• processing, inspection, transport, waiting and storage activities;
• decision points, rework loops and possible failure points;
• information flows, approvals and records; and
• measures associated with time, quality, cost and delivery.

The map should distinguish value-adding work from necessary but non-value-adding work and from avoidable waste. Machining a required surface is value-adding; moving a component between distant departments is usually necessary but non-value-adding; correcting an incorrectly assembled component is avoidable waste. This distinction helps managers examine total flow rather than merely local labour utilization.

A process map is not automatically a standard operating procedure. It provides a structural view of the process. Detailed instructions, control limits, inspection methods and safety requirements belong in associated documentation. At postgraduate level, the important insight is that process mapping is both descriptive and diagnostic: it reveals where information, material and accountability are interrupted.

Bills of materials: product structure and planning logic

A bill of materials (BOM) is a formal representation of the components and quantities required to make, assemble or package a product. It is a product structure, not merely a purchasing list. A single-level BOM lists immediate components; a multilevel BOM shows parent-child relationships across several levels.

Consider a product P consisting of one subassembly A and two units of component B. A consists of two units of component C and one unit of component D. The multilevel structure is:

P (level 0)

• A (level 1): quantity 1
• C (level 2): quantity 2
• D (level 2): quantity 1
• B (level 1): quantity 2

To produce 100 units of P, gross requirements are 100 units of A, 200 units of B, 200 units of C and 100 units of D, before accounting for inventory, scheduled receipts, scrap and lot-sizing rules.

Important BOM attributes include item identification, description, unit of measure, quantity per assembly, effectivity date, engineering revision, scrap factor, phantom designation and supply type. A phantom item represents a logical grouping that may not be stocked independently. Engineering and manufacturing BOMs may differ because design describes product form, whereas manufacturing describes how the product is assembled and supplied.

BOM accuracy has direct consequences for material requirements planning, inventory, purchasing, product costing and traceability. Common errors include duplicate part numbers, incorrect units of measure, omitted consumables, outdated revisions and failure to reflect yield losses. Change control is essential: an engineering change should identify the affected item, effective date, disposition of existing stock and responsible approval authority.

Routings and work-centre data

A routing specifies the sequence of operations through which a part or assembly passes. Each routing operation normally includes an operation number, description, work centre, required tools or fixtures, setup requirements, run time, inspection requirements and queue or move assumptions. Alternative routings may be authorized where different machines or methods can produce an equivalent result.

The routing converts product structure into a time-phased manufacturing path. If a shaft must be cut, turned, heat-treated, ground and inspected, the sequence matters because it determines material release, capacity loading, work-in-process location and expected completion time. An incorrect sequence can create quality problems even when all individual operations are technically capable.

A work-centre record describes a capacity resource or group of similar resources. Typical fields include available machines, labour skills, shifts, calendars, planned downtime, efficiency factors, setup capability, tooling, subcontracting options and cost rates. Capacity may be expressed in machine hours, labour hours, standard hours, kilograms, litres or units per period. The choice should reflect the constraint that actually limits output.

Theoretical capacity is the maximum capacity under ideal conditions. Demonstrated or effective capacity reflects realistic losses such as breaks, maintenance, changeovers, absenteeism, quality losses and product mix. Confusing these concepts leads to systematically overloaded schedules. Capacity data should be reviewed when shifts change, equipment is modified, staffing changes or actual performance persistently differs from standards.

Time standards and the measurement of work

A time standard is an approved estimate of the time required by a qualified operator, using a specified method and equipment, to complete an operation under defined conditions. It may include setup time, run time, inspection time, allowance for personal needs and fatigue, and sometimes delay allowances. The precise composition must be stated because different organizations use “standard time” differently.

Several methods are available. Direct time study uses repeated observations with a stopwatch or digital system. Predetermined motion-time systems estimate work from defined human motions. Work sampling estimates the proportion of time spent on different activities through random observations. Historical data can be useful when products and methods are stable, but it may preserve inefficiencies. Standard data systems use previously validated elemental times for recurring tasks.

Observed time is not necessarily standard time. A basic formulation is:

Standard time = Normal time × (1 + allowance factor)

Normal time adjusts observed time for the assessed pace of the operator. Allowances recognize fatigue, personal needs and unavoidable delays. For example, if observed time is 8 minutes, the performance rating is 110%, and the allowance is 15%, normal time is 8.8 minutes and standard time is 10.12 minutes.

Standards must be method-specific. A change in tooling, batch size, material, inspection requirement or layout may invalidate an existing standard. Standards should not be used to pressure workers into unsafe speeds or to conceal chronic equipment failures. Their legitimate uses include capacity calculation, scheduling, labour planning, costing and improvement analysis.

Performance architecture: interpreting the main measures

Performance measures are meaningful only when their definitions, denominators, time horizons and data sources are explicit. A measure can be mathematically correct yet managerially misleading.

Productivity

Productivity relates output to input. Labour productivity may be expressed as acceptable units divided by labour hours, while multifactor productivity relates output to a combination of labour, materials, energy and capital. Productivity rises when the same input produces more acceptable output, or when the same output requires fewer inputs. It is not identical to worker effort: poor methods, unreliable equipment and material shortages may reduce productivity without reflecting individual performance.

Utilization and efficiency

Utilization measures the extent to which available capacity is occupied. A common expression is:

Utilization = Actual productive time ÷ Available scheduled time × 100

Efficiency compares actual performance with a standard or planned reference. One expression is:

Efficiency = Standard hours earned ÷ Actual hours worked × 100

A work centre can have high utilization and low efficiency if it is busy producing slowly, reworking output or processing an unfavorable product mix. Conversely, low utilization and high efficiency may indicate that operators work quickly when jobs are available but suffer from insufficient demand or poor scheduling.

Throughput

Throughput is the rate at which the system produces and sells, or otherwise completes, acceptable output. In a constrained system, improving a non-bottleneck may not increase total throughput. The theory of constraints therefore directs attention to identifying the constraint, exploiting it, subordinating other decisions to it, elevating its capacity and repeating the process when the constraint moves.

Lead time and cycle time

Lead time is the elapsed time from the beginning of a request or order to its completion or delivery. Manufacturing cycle time usually refers to the elapsed time required to make a product through the production process. Processing time is only one component. Lead time may include order entry, planning, queueing, setup, processing, inspection, transport, waiting for material and dispatch.

A useful relationship is Little’s Law:

Work-in-process = Throughput rate × Flow time

The relationship applies under stable conditions and consistent units. It shows why reducing unnecessary work-in-process can reduce flow time, and why increasing WIP without increasing throughput often increases congestion and lead time.

Service and delivery performance

Service measures translate operational performance into customer consequences. On-time delivery may be measured against the confirmed date, requested date or promised date; the chosen definition must be consistent. Other measures include fill rate, order completeness, perfect-order performance, response time and schedule adherence. High average on-time delivery can conceal severe failures for strategically important customers, so segmentation and distributional analysis are valuable.

Cost measures

Production cost may include direct material, direct labour, variable overhead, fixed overhead, subcontracting, quality costs, inventory carrying cost and logistics. Standard cost supports planning and variance analysis, but it is not necessarily the cash cost of a particular decision. Managers should distinguish price, quantity, usage, rate, volume, mix and efficiency variances and investigate their operational causes rather than treating variance as a verdict on a department.

Integrating data with planning and control

The normal information flow begins with demand and product definition. The BOM determines material requirements; the routing determines operation requirements; work-centre records determine available capacity; and time standards support loading and completion estimates. The master production schedule then translates aggregate demand into planned production, while material and capacity planning test feasibility.

At execution level, dispatch lists prioritize operations, shop-floor reporting records actual start and finish times, inventory transactions update material status, and quality records identify accepted, rejected and reworked quantities. Feedback from actual performance should improve standards and master data, but changes require governance. Uncontrolled edits can make historical comparisons meaningless and create disputes about accountability.

A useful performance dashboard balances leading and lagging measures. Schedule adherence and queue length are leading indicators of future delivery risk; completed output, scrap and cost are lagging indicators. Measures should be reviewed at the level where action is possible, while preserving an end-to-end view of customer value.

Digital systems, analytics and emerging practice

Enterprise resource planning systems, manufacturing execution systems and product lifecycle management platforms increasingly share production data. Barcode and RFID transactions improve traceability; machine connectivity supplies runtime and downtime data; and digital work instructions can present revision-controlled methods at the point of use. Advanced planning systems can simulate alternative schedules and capacity scenarios.

Analytics can compare standard and actual times, identify recurring bottlenecks, detect unusual downtime and estimate the effect of schedule changes. Machine-learning models may support predictive maintenance or anomaly detection, but their results depend on representative data and stable definitions. Automation does not eliminate the need for master-data governance. An algorithm trained on incorrect routings can produce a highly optimized but infeasible schedule.

Current practice is moving toward digital twins, real-time production monitoring, integrated quality data and event-based planning. The managerial challenge is to combine speed of information with disciplined interpretation. Real-time visibility is valuable only when responsibilities, escalation rules and corrective methods are clear.

Illustrative case: improving the delivery performance of Orion Pumps

Background

Orion Pumps, a fictional medium-sized manufacturer, produces industrial pumps in low-volume batches. Its planning system contains BOMs, routings and work-centre calendars, but data ownership is unclear. Delivery performance has declined, overtime has increased and supervisors frequently expedite jobs.

Situation and problem

The monthly report shows machining utilization of 96%, assembly utilization of 82% and overall on-time delivery of 68%. Management concludes that machining needs another machine. However, customer orders spend an average of four days waiting between machining operations, while actual machining time averages only six hours. Inspection also reports repeated rework on a critical shaft dimension.

A data audit finds that machining standards exclude a recurring fixture-change activity, the routing does not show a required inspection queue, and the BOM contains an obsolete seal revision. The work-centre calendar assumes 480 minutes per shift although planned maintenance removes 45 minutes. Operators also record completed quantities before final inspection.

Analysis

The high utilization figure reflects a denominator that overstates available time and includes work that is waiting for inspection. The apparent capacity shortage is partly a data and flow problem. The obsolete BOM causes material substitutions and additional approval time. The missing fixture activity makes schedules optimistic, while premature completion reporting inflates throughput and masks rework.

Approach and solution

Orion establishes a cross-functional master-data team involving engineering, production, quality, planning and finance. It revises the BOM with effectivity dates, documents the routing including inspection and fixture operations, and recalculates work-centre calendars using actual planned availability. A time study creates separate setup and run standards. Completed output is recognized only after the defined quality release.

The team then identifies machining as the constraint during peak weeks. It sequences similar shaft families together to reduce setup frequency, protects a small time buffer before the constraint, and schedules preventive maintenance outside the constrained window. A dashboard reports accepted throughput, queue time, first-pass yield, schedule adherence and constraint utilization rather than utilization alone.

Outcome

In this illustrative case, the principal improvement comes not from immediately purchasing equipment but from restoring data integrity and controlling flow. Management can now distinguish genuine capacity loss from waiting, rework and incorrect calendar assumptions. The revised information also provides a sound basis for deciding whether additional capacity is eventually justified.

Learning points

The case demonstrates that performance metrics are constructed from definitions and data structures. A high utilization value may coexist with poor delivery; a routing error may appear as a labour problem; and a BOM error may appear as a purchasing problem. Integrated diagnosis is therefore more reliable than isolated departmental reporting.

Risks, limitations and governance concerns

The most common mistake is to treat master data as permanent. Products, processes, suppliers, equipment and regulations change. Other errors include using average times for highly variable products, ignoring queue and transport time, comparing unlike work centres, rewarding local utilization, and counting defective output as productive output.

Performance systems can also create dysfunctional behaviour. If supervisors are rewarded only for machine utilization, they may produce unwanted inventory. If employees are judged solely against standards, they may avoid difficult jobs or suppress problem reporting. If on-time delivery is pursued without quality controls, expedited and reworked output may increase total cost.

Governance should specify data owners, approval rights, revision control, audit trails, naming conventions, effective dates and periodic validation. Ethical practice requires accurate reporting, respect for worker safety and privacy, non-discriminatory performance evaluation and transparent use of monitoring technologies. Regulatory requirements may also affect traceability, calibration, product records and retention of production data.

Best-practice principles

Effective organizations generally:

1. define every metric, denominator, time window and data source;

2. assign ownership for BOMs, routings, standards, calendars and cost rates;

3. separate setup, run, queue, move, inspection and delay times;

4. measure accepted output and make quality visible in throughput;

5. validate standards after method, product or equipment changes;

6. balance local efficiency with end-to-end flow and customer service;

7. use exception-based dashboards rather than excessive reporting;

8. integrate engineering, operations, quality, supply chain and finance; and

9. review master data through controlled, evidence-based change processes.

Chapter Summary

Production planning and control depends on a coherent architecture of product, process and resource data. Process maps show how work and information flow; BOMs define product structure and material requirements; routings define operation sequences; work-centre records describe capacity; and time standards support scheduling, costing and performance analysis. Productivity, utilization, efficiency, throughput, lead time, service and cost measures answer different questions and must not be interpreted in isolation. Little’s Law and constraint-based reasoning help connect flow and capacity decisions. Reliable governance, revision control and ethical measurement are essential, while digital systems and analytics extend visibility without replacing managerial judgment. The Orion Pumps case illustrates that poor delivery may arise from inaccurate data and flow design rather than from a simple shortage of equipment.

Key Terms

Bill of materials: A structured listing of components, quantities and relationships required for a product.

Capacity: The feasible output or resource time available during a specified period.

Cycle time: Elapsed time associated with producing a unit or completing a production cycle.

Efficiency: The relationship between standard performance and actual resource use.

Effectivity date: The date from which a particular product, BOM or process revision applies.

Lead time: Elapsed time from initiation of a request or order to completion or delivery.

Little’s Law: The relationship stating that WIP equals throughput multiplied by flow time under stable conditions.

Master data: Controlled, relatively stable information used repeatedly by planning and execution systems.

Process map: A visual representation of activities, decisions, flows and relationships in a process.

Productivity: Output produced per unit of input.

Routing: The ordered sequence of operations and resources used to manufacture a product.

Standard time: Approved time for a qualified operator to perform a defined task under specified conditions.

Throughput: The rate at which acceptable output is completed or sold.

Utilization: The proportion of available scheduled capacity that is occupied or used.

Work centre: A machine, labour group or resource category where operations are performed.

Work-in-process: Items that have entered production but are not yet completed.

First-pass yield: The proportion of output accepted without rework.

Schedule adherence: The extent to which actual production follows the planned schedule.

Review Questions

Short Answer Questions

1. Why is production data considered an operational capability rather than a clerical record?

2. Distinguish a single-level BOM from a multilevel BOM.

3. What information is normally contained in a routing?

4. Define utilization and efficiency and explain one difference between them.

5. What is the managerial significance of Little’s Law?

Long Answer Questions

1. Explain how BOMs, routings, work-centre records and time standards interact in production planning and control.

2. Discuss the methods used to develop time standards and the conditions under which each may be appropriate.

3. Explain why high utilization may coexist with poor customer delivery performance.

4. Describe a governance system for maintaining accurate production master data.

5. Discuss how digital systems and analytics can improve production control while creating new risks.

Analytical / Application Questions

1. A work centre has 420 scheduled minutes, produces 35 accepted units and has a standard of 10 minutes per unit. It records 390 minutes worked. Calculate utilization using productive time of 350 minutes and calculate efficiency using standard hours earned divided by actual hours worked. Interpret the results.

2. A factory increases WIP from 500 to 800 units while throughput remains 100 units per day. Using Little’s Law, estimate the change in average flow time and explain the likely operational implications.

3. A plant reports 98% machine utilization but 70% on-time delivery and rising rework. Using the chapter’s performance architecture, propose a diagnostic sequence and identify at least four data items that should be validated.

22. Multiple Choice Questions with Answers

Question 1

Which production record primarily defines the parent-child relationships and quantities of components in a product?

A. Routing

B. Bill of materials

C. Work-centre calendar

D. Dispatch list

Correct Answer: B. Bill of materials

Question 2

A work centre has high utilization but low efficiency. Which interpretation is most appropriate?

A. It is necessarily producing at low total cost

B. It is busy, but actual performance is below the relevant standard

C. It has no demand and no scheduled work

D. Its output must be entirely defect-free

Correct Answer: B. It is busy, but actual performance is below the relevant standard

Question 3

Observed time for an operation is 8 minutes, the performance rating is 110%, and the allowance is 15% of normal time. What is the standard time?

A. 8.00 minutes

B. 8.80 minutes

C. 9.20 minutes

D. 10.12 minutes

Correct Answer: D. 10.12 minutes

Question 4

Which item is most clearly a routing attribute rather than a BOM attribute?

A. Quantity of seals per pump

B. Engineering revision of a component

C. Sequence of machining and inspection operations

D. Unit of measure for a purchased part

Correct Answer: C. Sequence of machining and inspection operations

Question 5

Under stable conditions, throughput is 50 units per day and average flow time is 4 days. What is the implied average WIP according to Little’s Law?

A. 12.5 units

B. 54 units

C. 200 units

D. 250 units

Correct Answer: C. 200 units

Question 6

Why should accepted output rather than merely completed output be used when measuring effective throughput?

A. It excludes all labour costs

B. It prevents defective or unreleased items from being counted as useful output

C. It always increases reported productivity

D. It eliminates the need for quality inspection

Correct Answer: B. It prevents defective or unreleased items from being counted as useful output

Question 7

In the Orion Pumps case, which finding directly made the machining schedule overly optimistic?

A. Assembly utilization was 82%

B. The BOM contained an obsolete seal revision

C. Fixture-change activity was excluded from the machining standard

D. Operators used digital work instructions

Correct Answer: C. Fixture-change activity was excluded from the machining standard

Question 8

Which action best reflects constraint-based production management?

A. Maximizing output at every non-bottleneck regardless of demand

B. Protecting and subordinating other activities to the system constraint

C. Increasing WIP before every work centre without analysis

D. Using theoretical rather than effective capacity for scheduling

Correct Answer: B. Protecting and subordinating other activities to the system constraint

Question 9

Which governance practice most directly supports traceability of engineering and manufacturing changes?

A. Removing revision numbers from records

B. Allowing informal edits by all users

C. Using effectivity dates, approvals and audit trails

D. Replacing standards whenever a single delay occurs

Correct Answer: C. Using effectivity dates, approvals and audit trails

Question 10

A company rewards only machine utilization and employees produce items that are not currently needed. Which problem does this illustrate?

A. Local optimization conflicting with overall flow

B. Improved perfect-order performance

C. Accurate capacity modelling

D. Reduction of unnecessary inventory

Correct Answer: A. Local optimization conflicting with overall flow