Hill A.V. The Encyclopedia of Operations Management: A Field Manual and Glossary of Operations Management Terms and Concepts

Подождите немного. Документ загружается.

ptg6843605

cost center – cost of quality

The Encyclopedia of Operations Management Page 86

( 2) / (1 )t r n r  

has a Student’s t-distribution (approximately) with n − 2 degrees of freedom and can be

used to test if an r value is different from zero.

In Excel use CORREL(x_range, y_range), where x_range and y_range are ranges. CORREL will have an

error if the variance of x or y is zero. The equivalent formula (n/(n−1))*COVAR(x_range,

y_range)/STDEV(x_range)/STDEV(y_range) can also be used. The term n/(n−1) is needed because the Excel

COVAR function is for the population rather than the sample.

The correlation between two sets of ranked values (x

, y

) can be found with the Spearman’s rank correlation

coefficient, which is

2 2

1 ( ) ( 1))6 / (

i i

x y n n





   



. However, if two or more ranks are tied for either set of

numbers, the correlation coefficient (Pearson’s) should be used.

See autocorrelation, covariance, forecast error metrics, linear regression, variance.

cost center – An accounting term for an area of responsibility that is only held accountable for its cost.

Cost centers are often service and support organizations, such as manufacturing, human resources, and

information technology, that do not have any easily assignable revenue.

See absorption costing, Activity Based Costing (ABC), human resources, investment center, profit center,

revenue center.

cost driver – See Activity Based Costing (ABC).

cost of goods sold – An accounting term for all direct costs incurred in producing a product or service during a

period of time; also called cost of goods, cost of sales, cost of products sold, and cost of production.

Cost of goods sold usually includes direct materials, incoming transportation, direct labor cost, production

facilities, and other overhead (indirect) labor and expenses that are part of the manufacturing process. It does not

include indirect costs, such as administration, marketing, and selling costs, that cannot be directly attributed to

producing the product.

Inventory is an asset, which means it is not expensed when purchased or produced, but rather goes into an

inventory asset account. When a unit is sold, the cost is moved from the inventory asset account to the cost of

goods sold expense account. Cost of goods sold is on the income statement and used in the inventory turnover

calculation.

See ABC classification, direct cost, direct labor cost, financial performance metrics, gross profit margin,

inventory turnover, Last-In-First-Out (LIFO), overhead, transfer price.

cost of quality – A framework coined by quality leader Phillip Crosby and used to measure quality-related costs;

now called the “price of non-conformance.” 

The cost of quality concept was popularized by Phillip Crosby, a well-known author and consultant (Crosby

1979). Crosby focused on the following principles:

 Quality is defined as conformance to requirements.

 The system for causing quality is prevention, not appraisal.

 The performance standard is zero defects.

 The measurement of quality is the cost of quality (sometimes called the price of nonconformance).

More recently, Crosby has replaced “the cost of quality” with the “price of nonconformance” in response to

quality professionals who did not like the older term. The price of nonconformance assigns an economic value to

all waste caused by poor quality. Examples of the price of nonconformance include wasted materials, wasted

capacity, wasted labor time, expediting, inventory, customer complaints, service recovery, downtime,

reconciliation, and warranty.

According to Feigenbaum (1983), the cost of quality framework includes these four elements:

Prevention costs – Cost of designing quality into the product and process. This includes product design,

process design, work selection, and worker training. Some authors also add the cost of assessing and improving

process capability. Many firms find that this cost is hardest to measure.

Appraisal costs – Cost of inspection, testing, auditing, and design reviews for both products and procedures.

Internal failure costs – Cost of rework, scrap, wasted labor cost, wasted lost machine capacity, and poor

morale. Lost capacity for a bottleneck process can also result in lost gross margin.

ptg6843605

Council of Logistics Management (CLM) − Cpk

Page 87 The Encyclopedia of Operations Management

External failure costs – Costs incurred after the customer has taken possession of the product and include

warranty, repair (both field repair and depot repair), lost gross margin/refunds, customer support, lost customer

good will, damage to the brand, damage to channel partnerships, and lawsuits.

An important teaching point with this framework is that most organizations need to move the costs up the

list. In other words, it is usually better to have internal failures than external failures. It is usually better to have

appraisal than internal failure and usually better to have prevention than appraisal. A couple of metaphors are

helpful here. It is better to avoid smoking (prevention) than to try to heal cancer. It is better to avoid toxic waste

than to try to clean it up.

The “1-10-100 rule” suggests that $1 on prevention will save $10 in appraisal, $10 on appraisal will save

$100 in internal failure, and $100 in internal failure will save $1000 in external failure. Of course, the numbers

are not precise, but the concept is an important one.

The old saying “An ounce of prevention is worth a pound of cure” communicates the same concept. A

similar old phrase “A stitch in time saves nine” suggests that timely preventive maintenance will save time later.

In other words, sewing up a small hole in a piece of clothing will save time later.

See appraisal cost, early detection, inspection, prevention, process capability and performance, quality

management, scrap.

Council of Logistics Management (CLM) – See Council of Supply Chain Management Professionals

(CSCMP).

Council of Supply Chain Management Professionals (CSCMP) – A professional society with the mission

“to lead the evolving Supply Chain Management profession by developing, advancing, and disseminating supply

chain knowledge and research.”

Founded in 1963, the Council of Supply Chain Management Professionals (CSCMP) is an association for

individuals involved in supply chain management. CSCMP provides educational, career development, and

networking opportunities to its members. The National Council of Physical Distribution Management (NCPDM)

was founded in 1963 by a group of educators, consultants, and managers who envisioned the integration of

transportation, warehousing, and inventory as the future of the discipline. In 1985, the association changed its

name to the Council of Logistics Management (CLM) and in 2005 changed its name to CSCMP.

CSCMP publishes a number of trade journals and the academic journal The Journal of Business Logistics.

The CSCMP website is www.cscmp.org.

See operations management (OM).

counting tolerance – The margin for error used when counting items in an inventory.

An item count is considered wrong only when the count is off by more than the “counting tolerance,” which

is usually a percentage defined for each category of items. High-value items will usually have a counting

tolerance of zero; very low-value items might have a counting tolerance of 10% or more.

See cycle counting, tolerance.

covariance – A measure of the strength of association between two variables.

If two variables have high covariance, they tend to vary together. In other words, when one is higher (lower)

than its mean, the other one is too. The mathematical definition of the sample covariance between random

variables X and Y is

1 1 1 1

1 1 1

Cov( , ) ( )( )

1 1

n n n n

i i i i i i

i i i i

X Y x x y y x y x y

n n n

   

    

 

 

  

   

   

  

 

   

  

 

   

, where

and

are the sample means. The first term is called the definitional form and the last term is called the

computational form because it only requires a single pass for the summations. Covariance is related to

correlation by the equation

Cov( , ) / Var( )Var( )

x y

r x y x y

, which shows that correlation is the covariance

“normalized” by the standard deviations of the two variables. Some basic facts about covariance for random

variables X and Y and constants a and b: Cov(a,X) = 0, Cov(X,Y) = Cov(Y,X), Cov(X,X) = Var(X), Cov(aX, bY) =

abCov(X,Y), Cov(X + a, Y + b) = Cov(X,Y).

The Excel population covariance is COVAR(x_range, y_range), where x_range and y_range are the x and y

ranges. For the sample covariance, use (n/(n−1))COVAR. Note that this issue is not documented in Excel.

See Analysis of Variance (ANOVA), correlation, variance.

ptg6843605

Cp – Critical Path Method (CPM)

The Encyclopedia of Operations Management Page 88

– See process capability and performance.

CPFR – See Collaborative Planning Forecasting and Replenishment.

– See process capability and performance.

CPM – See Critical Path Method (CPM).

CRAFT – See facility layout.

crashing – See critical path.

critical chain – The set of tasks that determines the overall duration of a project, taking into account both

precedence and resource dependencies.

The critical chain is similar to the critical path except that it goes one major step further and factors in

resource constraints. The steps in the critical chain approach are as follows:

Step 1. Compute the early start, early finish, late start, late finish, and slack times as is normally done

with the critical path method – The path through the network with the longest total time is called the

critical path, which is the path (or paths) with the shortest slack. The estimated task times for this

process should be set at the median rather than the 90-99% point of the task time distribution.

Step 2. Create a detailed schedule, starting from the current date and moving forward in time – When a

task is complete, begin the next task in the precedence network if the required resources are available.

When a resource becomes available, assign it to the task that has the least amount of slack time as

computed in Step 1. Continue this process until the schedule is complete for all activities. This new

schedule is called the critical chain and will be longer than the critical path, which only considers the

precedence constraints. The critical chain will still follow the precedence constraints, but will never

have any resource used more than its capacity.

Step 3. Strategically add time buffers to protect activities on the critical chain from starting late – Non-

critical chain activities that precede the critical chain should be planned to be completed early so the

critical chain is protected from disruption.

The consulting/software firm Realization markets software called Concerto that implements critical chain

concepts in Microsoft Project.

See critical path, Critical Path Method (CPM), Earned Value Management (EVM), Project Evaluation and

Review Technique (PERT), project management, safety leadtime, slack time, Theory of Constraints (TOC).

critical incidents method – An approach for identifying the underlying dimensions of customer satisfaction.

The critical incidents method involves collecting a large number of customer (or worker) complaints and

compliments and then analyzing them to identify the underlying quality dimension (timeliness, friendliness, etc.).

It is important to note that in analyzing a complaint or a complement, it does not matter if it is a negative

(complaint) or a positive (compliment); the goal is to simply identify the underlying dimensions of quality,

regardless if they are satisfiers or dissatisfiers.

This technique is useful for identifying the key dimensions of service quality for a customer satisfaction

survey. This technique can also be used in other contexts, such as job analysis to identify the critical dimensions

of worker satisfaction.

See service quality, voice of the customer (VOC).

critical path – The longest path through a project-planning network; the path that has the least slack (float) time. 

The only way to reduce the project completion time is to reduce the task times along the critical path. Task

times on the critical path are usually reduced by applying more resources (people, money, equipment, etc.) or by

overlapping activities that were originally planned to be done sequentially. Reducing times along the critical

path is called crashing, schedule compression, and fast tracking. Crashing non-critical activities will not

improve the project completion time. Crashing should focus on the critical path tasks that have the lowest cost

per unit time saved.

The Critical Path Method entry explains the methodology for finding the critical path. Activities not on the

critical path can become critical if they are delayed. Therefore, project managers need to monitor schedules and

practice risk management. It is possible for a network to have two or more critical paths.

A critical chain is similar to a critical path, but the two concepts are not synonymous. The critical chain

considers resource constraints, whereas the critical path does not.

ptg6843605

Critical Path Method (CPM) − critical ratio

Page 89 The Encyclopedia of Operations Management

See critical chain, Critical Path Method (CPM), Failure Mode and Effects Analysis (FMEA), project

management, slack time.

Critical Path Method (CPM) – An approach for project planning and scheduling that focuses on the longest path

through the project planning network. 

Project scheduling begins with the work breakdown structure that defines the “tree” of activities that make

up the project. Each task at the bottom of the tree

is then defined in terms of a task name, task time,

resources (people, machines, money, etc.), and a

set of precedence relationships, which is the set

of tasks that need to be completed before the task

can be started.

The simple example to the right shows the

scheduling process with a project network with

five tasks (A-E) with the times (in days) noted by

each task. The process has two passes: a forward pass to create the early start and early finish times, and a

backward pass to create the last finish and late start times for each task.

The forward pass begins with task A, which has an early start at the beginning of day 1 (ES = 1) and an early

finish at the end of day 1 (EF = 1). Task B cannot start until task A is completed and therefore has an early start

at the beginning of day 2 (ES = 2). Task B requires 2 days and has an early finish at the end of day 3 (EF = 3).

The forward pass continues until every node is assigned an early start (ES) and early finish (EF).

The backward pass (back scheduling) begins with the desired completion date for the project, which is the

late finish for the last task (task E). The required completion date for this project is the end of day 11 (i.e., LF =

11). Planning backward from this completion date, task E has a late start date at the beginning of day 10 (LS =

10). Continuing this backward pass creates the late finish and late start for all nodes. The notation in parentheses

beside each task node is (Early Start, Late Start, Early Finish, Late Finish) = (ES, LS, EF, LF).

The slack time (float) for any task is the difference between the early start and the late start (i.e., LS – ES),

which is always the same as the difference between the early finish and late finish (i.e., LF – EF). The critical

path is the path through the network that has the smallest slack. In this example, the critical path is ADE.

The slack time for a task may be zero or negative.

Management should prioritize tasks along the critical path for both resource allocation and crashing. When a

resource (e.g., person, tool, or machine) becomes free, a good rule for re-allocating this resource is to use the

minimum slack rule, which simply assigns the resource to the open task that has the least slack. This is

essentially the same process used to identify the critical chain.

If it is necessary to reduce the total project time, the project manager should find the task on the critical path

that can be reduced (crashed) at the lowest cost per unit time, make the change, and then repeat the crashing

process until the desired project completion time is achieved or until the cost exceeds the benefit. Crashing non-

critical activities will not improve the project completion date.

Activities not on the critical path can become critical if they are delayed. This suggests that project managers

need to constantly monitor project schedules and practice good risk management to prepare for contingencies.

Note that it is possible for a project network to have two or more critical paths.

The above figure uses the “activity-on-node” approach. The alternative “activity-on-arc” approach is

presented in some textbooks, but is far more difficult to understand and is rarely used in practice.

Two of the better-known project scheduling packages are Microsoft Project and Primavera sold by Oracle.

Microsoft Project is said to be better for smaller projects and Primavera better for larger projects. Many other

commercial packages are available.

See back scheduling, critical chain, critical path, forward scheduling, Gantt Chart, load leveling, Project

Evaluation and Review Technique (PERT), project management, project network, slack time, work breakdown

structure (WBS).

critical ratio – See dispatching rule, newsvendor model.

(2, 5, 3, 6)

(4, 7, 6, 9)

(2, 4, 7, 9)

A,1

D,6

E,2

C,3 B,2

(ES, LS, EF, LF)

(8, 10, 9, 11)

(1, 3, 1, 3)

Simple CPM project network example

ptg6843605

Critical to Quality (CTQ) – cross-docking

The Encyclopedia of Operations Management Page 90

Critical to Quality (CTQ) – Key measurable characteristics of a product or process that require performance

standards or specification limits to satisfy the customer (internal or external) requirements; also called Key

Process Output Variable (KPOV).

CTQ may include the upper and lower specification limits or any other factors related to the product or

service. A CTQ usually must be interpreted from a qualitative customer statement to an actionable, quantitative

business specification. CTQs are what the customer expects of a product. The customer’s requirements must be

expressed in measurable terms using tools, such as DFMEA.

See Key Process Output Variable (KPOV), lean sigma, quality management, voice of the customer (VOC).

CRM – See Customer Relationship Management (CRM).

cross-functional team – A group of employees from different parts of an organization who come together and use

their different viewpoints and skills to address a problem.

Many organizational problems cannot be solved by a single business function. For example, a new product

development project might require expertise from marketing, sales, manufacturing, and engineering. Cross-

functional teams are often the best approach for addressing these problems. A cross-functional team may be self-

directed or directed by a sponsor (or sponsors) within the organization. Cross-functional teams are a common

component of concurrent engineering, agile software development, and lean sigma projects.

See agile software development, concurrent engineering, lean sigma, organizational design, project

management, Quality Function Deployment (QFD), red tag, sponsor.

cross-docking – A warehousing term for the practice of moving products directly from incoming trucks or rail cars

to outgoing trucks without placing inventory on shelves in the warehouse or distribution center; also called

crossdocking and cross docking.

Products that are good candidates for cross-docking have high demand, standardized packaging, and no

special handling needs (e.g., security or refrigeration). This is a common strategy for retail distribution where

trucks carry large shipments from factories to the cross-dock facility, which loads other trucks with mixed

assortments to send to retail stores.

Cross-docking has many advantages over traditional warehouse facilities:

• Reduced inventory and carrying cost – The firm replaces inventory with information and coordination.

• Reduced transportation cost – For less than truck load (LTL) and small package carriers, cross-docking is a

way to reduce transportation costs by consolidating shipments to achieve truck load quantities.

• Reduced labor cost – Cross-docking avoids costly moves to and from shelves in the warehouse.

• Improved customer service – Cross-docked shipments typically spend less than 24 hours in a cross-dock.

The figure on the right shows a cross-docking process with a top-down

view of eight trucks. Four trucks bring products in from suppliers and four

take products out to the retail stores.

Napolitano (2000) proposed the following classification scheme for

cross-docking:

Manufacturing cross-docking for receiving and consolidating inbound

supplies to support Just-in-Time manufacturing. For example, a manufacturer

might lease a warehouse close to its plant and use it to prep subassemblies or

consolidate kits of parts. Because demand for the parts is known, say from

the output of an MRP system, there is no need to maintain stock.

Distributor cross-docking for consolidating inbound products from

different vendors into a multi-SKU pallet, which is delivered as soon as the

last product is received. For example, computer distributors often source

components from different manufacturers and consolidate them into one

shipment in merge-in-transit centers before delivering them to the customer.

Transportation cross-docking for consolidating shipments from different shippers in the LTL and small

package industries to gain economies of scale. For small package carriers, material movement in the cross-dock

is by a network of conveyors and sorters; for LTL carriers it is mostly by manual handling and forklifts.

Retail cross-docking for receiving product from multiple vendors and sorting onto outbound trucks for

different stores. Cross-docking is one reason Walmart surpassed Kmart in retail sales in the 1980s.

Trucks coming in from suppliers

Trucks going out to retailers

ptg6843605

cross-selling − Croston’s Method

Page 91 The Encyclopedia of Operations Management

Opportunistic cross-docking is transferring an item directly from the receiving dock to the shipping dock to

meet a known demand in any type of warehouse.

All of the cross-docking practices involve consolidation and short cycle times, usually less than a day. Short

cycle time is possible because the destination for an item is known before or upon receipt.

See Advanced Shipping Notification (ASN), consolidation, distribution center (DC), distributor, dock,

economy of scale, forklift truck, less than truck load (LTL), logistics, Over/Short/Damaged Report, receiving,

supply chain management, Transportation Management System (TMS), warehouse, Warehouse Management

System (WMS).

cross-selling – A sales and marketing term for the practice of suggesting related products to a customer during a

sales transaction or a customer service encounter.

In contrast to cross-selling, up-selling is a sales technique where a salesperson attempts to persuade the

customer to purchase better or more expensive items. Cross-selling and up-selling can sometimes be combined

to increase the value added for the customer and also increase the sales and margin for the selling organization.

See call center, Customer Relationship Management (CRM), customer service, order entry.

cross-training – Training workers in several different areas or functions outside their normal job responsibilities.

Having workers learn a wide variety of tasks has many advantages:

Increased flexibility – Workers can provide backup when the primary worker is unavailable or when the

demand exceeds the capacity. This makes it easy to improve flow and reduce inventory. This increased

flexibility allows the line workers to dynamically balance the line without any help from an industrial engineer.

Process improvement – When workers have a broader understanding of the organization, they can be more

knowledgeable about how to improve it.

Develops human capital – Cross-trained workers are more valuable to the company and often find more

satisfaction in their jobs. Cross-training is often an investment in the future for a firm.

See cellular manufacturing, facility layout, human resources, job design, job enlargement, job rotation, lean

thinking, learning organization, line balancing, on-the-job training (OJT), workforce agility.

Croston’s Method – A time series forecasting method that is used for lumpy (intermittent) demand.

When the demand (sales) history for an item has many periods with zero demand, it is said to have

intermittent or lumpy demand. Exponential smoothing does a poor job of forecasting lumpy demand because

the exponentially smoothed average will approach zero after several zero demand periods and the forecast for the

period with the next “demand lump” will therefore be close to zero. Exponential smoothing, therefore, should

not be used for lumpy demand. A demand time series is often judged to be lumpy when the coefficient of

variation of the demand is greater than one.

Croston (1972) suggests a method for forecasting lumpy demand by decomposing the time series into (1) the

average size of the demand lumps and (2) the average time between the demand lumps. Croston’s Method uses

simple exponential smoothing (SES) for the average size of the demand lumps and also for the average number

of periods between the demand lumps. As shown in the table below, it implements three recursive equations.

The first equation updates the smoothed average size of the non-zero demand z

. If the demand in period t is

zero (i.e., d

= 0), the average size of the demand remains unchanged; otherwise, the average size of a non-zero

demand (z

) is updated with simple exponential smoothing. The second equation updates the average number of

periods since the last non-zero demand (p

) based on q

, which is the number of periods since the last demand.

When the demand is zero, p

remains unchanged; otherwise, p

is updated with simple exponential smoothing

based on q

. The third equation updates q

, which counts the number of periods between non-zero demand. If d

is zero, the number of periods since the last demand is incremented by 1; otherwise, the number of periods since

the last demand is reset to 1.

Croston’s Method rules

Average size of a

non-zero demand (z

)

Average number of periods since

the last non-zero demand (p

)

Number of periods since

the last demand (q

)

if 0

(1 ) if 0

t t

t t t

z d

d z d

 





  







if 0

(1 ) if 0

t t

t t t

p d

q p d

 







 







1 if 0

t t

q d



 











ptg6843605

CRP (Capacity Requirements Planning) – customer leadtime

The Encyclopedia of Operations Management Page 92

The forecast for the demand in period t + n is then

t n t t

z p





. If every period has a non-zero demand (i.e.,

= 1), Croston’s Method is equivalent to simple exponential smoothing with smoothing constant α.

The SAP Library (http://help.sap.com/saphelp_scm50/helpdata/en/ac/216b89337b11d398290000e8a49608/

content.htm) recommends initializing {z

, p

} = {0, 2} if d

= 0 and {d

, 1} if d

> 0. However, this will create

forecasts f

= 1 for all periods when no non-zero demand is encountered. Therefore, the values z

= 0 if d

= 0

and z

= d

if d

> 0 are more logical. The q

parameter should be initialized at 1.

An alternative approach is to increase the size of the time buckets so that zero demands are rare or use the

average demand over several periods. It is important to avoid using exponential smoothing for intermittent

demand because it will forecast relatively high demand right after a “lump” in demand and forecast nearly zero

demand about the time the next “lump” in demand occurs.

See coefficient of variation, exponential smoothing, forecasting, lumpy demand, time bucket, time series

forecasting.

CRP (Capacity Requirements Planning) – See Capacity Requirements Planning (CRP).

CTO – See configure to order (CTO).

CTQ – See Critical to Quality (CTQ).

cube utilization – A warehousing term for the percent of the usable three-dimensional space (volume) used in a

trailer, container, or warehouse.

Volume is calculated as the product of the three dimensions – width x height x depth.

See operations performance metrics, shipping container, slotting, trailer, warehouse.

cumsum chart – See cumulative sum control chart.

Cumulative Distribution Function (CDF) – See probability density function.

cumulative leadtime – The critical path leadtime (longest) required to purchase material and create a product to

offer to the market; also known as the stacked leadtime.

The cumulative leadtime usually is the time required to purchase materials, fabricate (cut, mold, weld, finish,

etc.), assemble, test, package, and ship a product assuming that no inventory is on hand and no materials have

been ordered. The customer leadtime might be much less than the cumulative leadtime if intermediate products

(e.g., subassemblies) are inventoried.

See customer leadtime, leadtime, planning horizon, push-pull boundary, time fence.

cumulative sum control chart – A quality and industrial engineering term for a graphical tool (chart) that plots

the cumulative sum of deviations over time; also known as a cumsum control chart or cumsum chart.

A cumulative sum control chart plots the cumulative sum of deviations of successive samples from a target

value (i.e.,

( )

m i

S x





 



), where S

is the cumulative sum, x

is the observed value (or sample mean), and μ is

the target value. This can also be defined recursively as

1m m i

S S x







 

. This concept is very similar to a

tracking signal used for forecasting.

See control chart, Statistical Process Control (SPC), Statistical Quality Control (SQC), tracking signal.

current reality tree – A Theory of Constraints term for a causal map that describes the current state of a problem.

See causal map, future reality tree, Theory of Constraints (TOC).

Customer Effort Score (CES) – A measure of service quality using the single question “How much effort did

you personally have to put forth to handle your request?”

Dixon, Freeman, and Toman (2010) developed this simple scale for call centers, service centers, and self-

service operations. They argue that customers do not want to be delighted; they just want their needs met

without annoying phone calls, transfers, and confusion. They claim that the CES is more predictive of customer

loyalty than the Net Promoter Score (NPS) and direct measures of customer satisfaction. They recommend using

a CES scale from 1 (very low effort) to 5 (very high effort) and suggest that service providers intervene when

customers report a high CES. Note that CES is a measure of dissatisfaction and NPS is a measure of satisfaction.

See Net Promoter Score (NPS), operations performance metrics, service management, service quality.

ptg6843605

customer leadtime − cycle counting

Page 93 The Encyclopedia of Operations Management

customer leadtime – The planned or actual time in system for a customer (or customer order); planned or actual

turnaround time. 

Customer leadtime is the difference between the time the customer (or order) enters the system and the time

that the customer (or order) is complete. The term “leadtime” suggests that this term should refer to a planning

factor rather than the actual time or the historical average leadtime.

In a manufacturing context, the customer leadtime begins after the push-pull boundary, which is the

location in the process where the firm no longer stocks inventory. Customer leadtime can be nearly zero if the

manufacturer has the push-pull boundary at finished goods inventory. In a retail context, customer leadtime is

the difference between the time the customer selected and received an item.

See addition principle, assemble to order (ATO), cumulative leadtime, leadtime, postponement, push-pull

boundary, respond to order (RTO), time in system, turnaround time.

customer profitability – The revenue a customer generates minus the costs needed to acquire and retain that

customer.

Customer profitability is closely related to the concepts of customer equity and lifetime value. Without a

method for estimating customer profitability, a firm may spend scarce marketing dollars to retain its unprofitable

(or low profit) customers and may mistreat its most profitable customers.

See Activity Based Costing (ABC), Customer Relationship Management (CRM).

Customer Relationship Management (CRM) – An information system that collects data from a number of

customer-facing activities to help an organization better understand its customers so that it can better match its

products and services to customer needs and thereby increase sales.

Although CRM involves information technology, it is fundamentally a strategic process (and not just an IT

project) for helping organizations better understand their customers’ needs, better meet those needs, and increase

sales and profits. CRM processes and systems combine information from marketing, sales, contact management,

and customer service activities. CRM also provides tools to analyze customer/product sales history and

profitability, campaign tracking and management, contact and call center management, order status information,

and returns and service tracking. A good CRM system provides many benefits:

• Provide exactly the services and products that customers want.

• Offer better customer service.

• Allow for more effective cross-seling (selling complementary products).

• Help sales staff close deals faster.

• Help the firm retain current customers and discover new ones.

• Collect timely complete information on customers through multiple customer interfaces, including call

centers, e-mail, point-of-sale operations, and direct contact with the sales force.

• Reduce the transaction cost for buying products and services.

• Provide immediate access to order status.

• Provide support that will reduce the costs of using products and services.

• Help management develop a deeper understanding of customer buying behavior “sliced and diced” in a

number of ways, such as geography, demographics, channel, etc.

See cross-selling, customer profitability, customer service, target price, voice of the customer (VOC).

customer satisfaction – See service quality.

customer service – (1) The organization charged with dealing with customer complaints and other customer needs.

(2) A metric used to evaluate how well an organization or an inventory is meeting customer requirements.

See Automatic Call Distributor (ACD), call center, cross-selling, Customer Relationship Management

(CRM), empowerment, fulfillment, order entry, service level, service quality, SERVQUAL, supplier scorecard,

Third Party Logistics (3PL) provider.

customization flexibility – See flexibility.

customization point – See push-pull boundary.

cycle counting – A methodology for counting items in storage that counts more important items more often and

systematically improves the record-keeping process. 

ptg6843605

cycle service level – cycle time

The Encyclopedia of Operations Management Page 94

Instead of counting all items with a year-end physical inventory count, cycle counting counts items

throughout the year, with the “important” items counted much more often than other items. Cycle counting is an

application of Pareto’s Law, where the “important few” items are counted often and the “trivial many” items are

counted infrequently.

The benefits of a good cycle counting program over a physical inventory are summarized in the table below.

The main benefits of cycle counting are (1) it is better at finding and fixing the problems that cause inaccurate

inventory balances and (2) it maintains record accuracy throughout the year.

Year-end inventory versus cycle counting

Year-end physical inventory Cycle counting

Frequency Once per year Daily or weekly for very important items;

less frequently for less important items

Correct balance Once per year Almost always throughout the year

Fix root problems No, only fixes the symptoms Yes

Improve improvement culture Very limited Very strong

Accountability for accuracy Weak Strong

Prioritizes counting effort No Yes

Disruption of production Major at year-end Minimal

Source: Professor Arthur V. Hill

Some factors to consider when determining how often to count an item include:

 Criteria set by accounting and external auditors for the elimination of the annual physical inventory count.

 External regulations that require specific count frequencies.

 Annual dollar volume of an item (also known as the annual cost of goods sold).

 Annual unit volume of an item (number of units sold).

 Unit cost of an item.

 Pilferage risk associated with the item.

 Current inventory accuracy level for that particular item.

Some rules for determining how often to count an item include (1) count items with higher annual dollar

volume more often (the ABC system), (2) count just before an order is placed, (3) count just before a new order

is placed on the shelf, (4) count when the physical inventory balance is zero, (5) count when the physical

inventory balance is negative, and (6) count after a specified number of transactions. Rules 2 and 3 are ways to

implement rule 1 because “A” items will be ordered more often. Rule 3 minimizes the number of units that need

to be counted because the shelf is nearly empty when an order arrives. Rules 2 and 3 can be implemented so that

every n-th order is counted. Rule 5 is a good rule to use along with any other rule.

An item count is considered wrong only when the count is off by more than the counting tolerance, which is

usually a percentage defined for each category of items. High-value items will usually have a counting tolerance

of zero; low-value items might have a counting tolerance of 10% or more.

With a blind count the counter is given the item number and location but no information about the count

currently in the database. This approach avoids giving counters reference points that might bias their counts.

Tompkins Associates provides more information on this topic at tompkinsinc.com/publications/monograph/

monographList/WP-19_Cycle_Counting.pdf?monographID=WP-19.

See ABC classification, backflushing, learning organization, on-hand inventory, Pareto’s Law, physical

inventory, scale count, shrinkage.

cycle service level – See safety stock.

cycle stock – The inventory due to lotsize quantities greater than one; also called the lotsize inventory. 

Cycle stock follows a “saw tooth” pattern (see the reorder point entry for an example). For instantaneous

replenishment and constant average demand, the average cycle stock is Q/2, where Q is the fixed order quantity.

Organizations can reduce cycle stock by reducing lotsizes and lotsizes can be reduced economically by reducing

the order (setup) cost. The setup time reduction methods entry provides a methodology for reducing setup cost.

See commonality, Economic Order Quantity (EOQ), instantaneous replenishment, lotsize, safety stock, setup

time reduction methods.

ptg6843605

cycle time − dampened trend

Page 95 The Encyclopedia of Operations Management

cycle time – (1) The time between completions (or starts) of a unit of work. (2) The time from beginning to end for

a unit of work (also known as throughput time, flow time, and turnaround time). 

These two definitions are quite different, but both are used in practice.

Definition 1: For decades, industrial engineers defined cycle time as the time between completions (or starts)

of a process step. For example, this could be measured as the time between units “falling off” the end of a

manufacturing process or the maximum work time allowed for each worker on an assembly line. In lean

terminology, the target cycle time (time between completions) aligned to the market demand rate is called the

takt time (see the entry for takt time). This is the definition for cycle time used in LEI’s Lean Lexicon

(Marchwinski & Shook 2006).

Definition 2: Cycle time can also be defined as the cumulative time (also called total throughput time, total

flow time, and production leadtime) required for a unit from start to end and can be measured as the completion

time minus the start time for the same unit.

The first definition is the time between completions and the second definition is the time to complete one

unit. The second definition (throughput time) has become the more widely used term in practice in North

America, and it is now common to use the terms throughput time (or flow time) and cycle time synonymously.

To compare the two definitions of cycle time, imagine an assembly line in an appliance factory that has one

dishwasher coming off the end of the line every two minutes. Each of the 20 steps in the assembly process

requires exactly two minutes. Using the first definition, this is a cycle time (time between completions) of two

minutes. However, using the second definition, a dishwasher that started at 8:00 am will not be complete until

8:40 am (20 steps times two minutes each), which means that the assembly line has a cycle time (throughput or

flow time) of 40 minutes. The first definition focuses on the time between units being completed for each step in

the process, whereas definition two focuses on the time to complete one unit from beginning to end for the entire

process.

Although it is fairly easy to measure the average cycle time (throughput time) for a manufacturing order in a

factory, it is often difficult to estimate the average cycle time (throughput time) for a complex product because it

is not clear when the product is started. Given that each required component might have a different starting time,

it is hard to determine the total cycle time for the completed product. A simple approach for estimating the cycle

time (and also the periods supply) is to use the inverse of the inventory turnover ratio for the work-in-process

(WIP) inventory. Therefore, if the WIP inventory turnover for a product is four turns per year, the “dollar

weighted cycle time” is three months.

In queuing theory terms, the average completion rate will be the same as the average production rate

(assuming that no units are lost and that sufficient capacity is available). With a production rate



(lambda), the

average time between starts and average time between completions is 1/



. According to Little’s Law (Little

1961), the average work-in-process inventory is





, where W

is the average time in the system. Building

on the previous example, the average time between completions was two minutes, which means that the

production rate is

1/ 2





units/minute. Given that the throughput time was

W 

minutes, the average

number of units in process is

s s

L W



  

= 20 units. See the queuing theory and Little’s Law entries for

more details on these issues.

Many practitioners use the terms “cycle time” (as in throughput time) and “leadtime” synonymously.

However, in this author’s opinion, it is better to use the term “leadtime” as a planning factor (as in planned

leadtime) rather than as the actual throughput time for an order. The leadtime entry discusses these issues.

See assembly line, balanced scorecard, inventory turnover, leadtime, Little’s Law, operations performance

metrics, order-to-cash, periods supply, pitch, purchasing leadtime, station time, takt time, time in system, touch

time, turnaround time, value added ratio, wait time.

cycle time efficiency – See value added ratio.