Hill A.V. The Encyclopedia of Operations Management: A Field Manual and Glossary of Operations Management Terms and Concepts
Подождите немного. Документ загружается.
ptg6843605
discriminant analysis – dispatching rules
The Encyclopedia of Operations Management Page 106
discriminant analysis – A statistical technique that predicts group membership; also called linear discriminant
analysis.
See linear regression, logistic regression.
diseconomy of scale – The forces that cause organizations to have higher unit costs as volume increases; also
called diseconomies of scale.
Most business managers are familiar with the concept of economy of scale, where the unit cost decreases as
the volume increases. The less-familiar concept of diseconomy of scale is where the unit cost increases as the
volume increases. In many industries, firms will have economies of scale until they grow to be quite large.
For example, it is said that the optimal hospital size is roughly 400 beds
10
and that larger hospitals tend to
become less efficient due to the complexity of the operation, which is a function of the number of employees and
the distance people have to travel. This is also true for high schools, where the optimal school size is probably
between 400 and 600 students
11
. A watch factory in Moscow once had 7,000 people making watches and a U.S.
defense factory was so large that workers had to ride bicycles to travel within the plant
12
. These factories were
not very competitive due to their size. The reasons why unit cost might increase with volume include:
Coordination and communication problems – As firm size increases, coordination and communication
becomes much more difficult and often leads to the creation of a large bureaucracy. The number of unique
pairs in an organization with n units or individuals is n(n − 1)/2, which grows with n
2
.
Top-heavy management – As firm size increases, management expense tends to increase.
Insulated managers – As firms increase in size, managers are less accountable to shareholders and markets
and are more insulated from reality. Therefore, they tend to seek personal benefits over firm performance.
Lack of motivation – As firms grow, workers tend to be more specialized, have a harder time understanding
the organization’s strategy and customers, and are more likely to be alienated and less committed to the firm.
Duplication of effort – As firm size increases, it is common for firms to waste money on duplicate efforts
and systems. It is reported that General Motors had two in-house CAD/CAM systems and still purchased
other CAD/CAM systems from outside firms.
Protection from consequences – In a small firm, most managers immediately see and experience the
consequences of their decisions. In many large firms, managers are transferred every few years and rarely
have to live with their bad decisions very long. Therefore, they do not learn from their mistakes.
Inertia – It is often very hard for a large organization to change directions. A VP of MIS for a large bank in
Seattle reported that it was nearly impossible for his bank to make any significant changes. It was just too
hard to understand all the linkages between the more than ten million lines of code.
Self-competition – The managing directors for a large paper products firm in Europe identified their biggest
problem as competition with the other operating companies within the same firm.
Transportation – If output for a national or international market is concentrated at a single large plant,
transportation costs for raw materials and finished goods to and from distant markets may offset scale
economies of production at the large plant.
Canbäck et al. (2006) found empirical support for many of the above statements. Some of the ideas above
were adapted from http://en.wikipedia.org/wiki/Diseconomies_of_scale.
See economy of scale, flexibility, pooling.
disintermediation – Removing a supplier or distributor from a supply chain; also called “cutting out the
middleman.”
10
The website www.ncbi.nlm.nih.gov/pmc/articles/PMC1116851 supports this statement.
11
The website www.wested.org/online_pubs/po-01-03.pdf supports this statement
12
These statements are based on personal experiences of the author.
ptg6843605
dispatch list − disruptive technology
Page 107 The Encyclopedia of Operations Management
A distributor is an intermediary between a manufacturer and its customers. When a firm removes a
distributor between it and its customers, it is said to have disintermediated the distributor and has practiced
disintermediation. This is common when a manufacturing firm replaces distributors with a website that sells
directly to customers. Reintermediation occurs when the distributor finds a way to re-insert itself into the
channel, possibly by offering its own website and better service.
See channel conflict, dot-com, supply chain management.
dispatch list – See dispatching rules.
dispatching rules – Policies used to select which job should be started next on a process; sometimes called job
shop dispatching or priority rules.
For example, a manager arrives at work on a Monday morning and has 20 tasks waiting on her desk. Which
task should she handle first? She might take the one that is the most urgent (has the earliest due date), the longest
one, or the one that has the most economic value. In a very similar way, a shop supervisor might have to select
the next job for a machine using the same types of rules.
The best-known dispatching rules include First-In-First-Out (FIFO), shortest processing time, earliest
due date, minimum slack time, and critical ratio, which are factors of the arrival time, processing time, due
date, or some combination of those factors. Other factors to consider include value, customer, and changeover
cost or time. The FIFO rule may be the “fairest” rule, but does not perform well with respect to average flow
time or due date performance. It can be proven that the shortest processing time rule will minimize the mean
(average) flow time, but does poorly with respect to on-time delivery. MRP systems backschedule from the due
date and therefore are essentially using a minimum slack rule, which has been shown to perform fairly well in a
wide variety of contexts. The critical ratio (Berry and Rao 1975) for a job is equal to the time remaining until the
due date divided by the work time remaining to complete the job
13
. A critical ratio less than one indicates the job
is behind schedule, a ratio greater than one indicates the job is ahead of schedule, and a ratio of one indicates the
job is on schedule. (Do not confuse the critical ratio rule with the critical ratio in the newsvendor model.)
Dispatching rules are used to create a daily dispatch list for each workcenter. This is a listing of
manufacturing orders in priority sequence based on the dispatching rule.
See expediting, First-In-First-Out (FIFO), heijunka, job shop, job shop scheduling, Last-In-First-Out
(LIFO), on-time delivery (OTD), operation, service level, shop floor control, slack time.
disruptive technology – A term coined by Professor Clayton Christensen (Christensen 1997; Christensen and
Raynor 2003) at Harvard Business School to describe a technological innovation, product, or service that
eventually overturns the existing dominant technology in the market despite the fact that the disruptive
technology is radically different than the leading technology and often has poorer performance (at least initially)
than the leading technology.
The disruptive technology often starts
by gaining market share in the lower price
and less demanding segment of the market
and then moves up-market through
performance improvements and finally
displaces the incumbent’s product. By
contrast, a sustaining technology provides
improved performance and will almost
always be incorporated into the
incumbent’s product.
In some markets, the rate at which products improve is faster than the rate at which customers can learn and
adopt the new performance. Therefore, at some point the performance of the product overshoots the needs of
certain customer segments. At this point, a disruptive technology may enter the market and provide a product
13
The critical ratio dispatching rule can be implemented in software as: If due > now then CR = (1 + due – now)/(1 + trpt);
otherwise CR = (due – now)/(1 + TRPT), where due is the due date (in hours), now is the current date (in hours), and trpt is the
total remaining processing time in hours (Rose, 2002). Note that other variants of the critical ratio rule are presented in the
literature and used in practice.
Examples of displaced and disruptive technologies
Displaced technology Disruptive technology
Canals
Railways
Photographic film
Traditional telephone systems
Vertically integrated steel mills
Disk drives
Railways
Automobile
Digital cameras
Voice over IP
Mini steel mills
Flash drives
ptg6843605
distinctive competence – distributor
The Encyclopedia of Operations Management Page 108
that has lower performance than the incumbent technology, but exceeds the requirements of certain segments,
thereby gaining a foothold in the market.
Christensen distinguishes between “low-end disruption” (that targets customers who do not need the full
performance valued by customers at the high end of the market) and “new-market disruption” (that targets
customers who could previously not be served profitably by the incumbent). The disruptive company will
naturally aim to improve its margin and therefore innovate to capture the next level of customers. The incumbent
will not want to engage in a price war with a simpler product with lower production costs and will move up-
market and focus on its more attractive customers. After a number of iterations, the incumbent has been
squeezed into successively smaller markets. When the disruptive technology finally meets the demands of its
last segment, the incumbent technology disappears. Some of the above information was adapted from
http://en.wikipedia.org/wiki/Disruptive_technology (November 16, 2006).
See market share, nanotechnology, New Product Development (NPD), technology road map.
distinctive competence – See core competence.
distribution – (1) In the logistics context, management of the movement of materials from the supplier to the
customer; also called physical distribution. (2) In the statistics context, a description of the range of values that a
random variable can attain and information about the probability of each of these values.
In the logistics context, distribution involves many related disciples, such as transportation, warehousing,
inventory control, material handling, and the information and communication systems to support these activities.
For information regarding the statistics context, see the probability distribution entry.
See distribution center (DC), distribution channel, distribution network, DRP, inventory management,
logistics, reverse logistics.
distribution center (DC) – A location used to warehouse and ship products.
See Advanced Shipping Notification (ASN), cross-docking, distribution, Distribution Requirements Planning
(DRP), distributor, logistics, warehouse, Warehouse Management System (WMS), wave picking.
distribution channel – The way that a product is sold and delivered to customers.
For example, a product, such as 3M sandpaper, might be sold through big box retailers, such as Home Depot,
or through national distributers, such as Granger. Other products might be sold through retail grocery stores
(e.g., Kroger), national grocery wholesalers (e.g., SuperValu), a company-employed national salesforce (e.g.,
IBM), or a company-owned Web-based channel (e.g., www.BestBuy.com). Firms in the channel are sometimes
called channel partners, even though technically the business relationships are rarely partnerships.
See big box store, channel conflict, channel integration, channel partner, distribution, distribution network,
distributor, logistics, supply chain management.
distribution network – The organizations, facilities, means of transportation, and information systems used to
move products from suppliers of suppliers to customers of customers.
See distribution, distribution channel, logistics.
Distribution Requirements Planning (DRP) – A planning system for managing inventory in a distribution
network; also called Distribution Resource Planning.
DRP is an extension of MRP for planning the key resources in a distribution system. According to
Vollmann, Berry, Whybark, and Jacobs (2004), “DRP provides the basis for integrating supply chain inventory
information and physical distribution activities with the Manufacturing Planning and Control system.”
DRP performs many functions such as:
Managing the flow of materials between firms, warehouses, and distribution centers.
Helping manage the material flows like MRP does in manufacturing.
Linking firms in the supply chain by providing planning records that carry demand information from
receiving points to supply points and vice versa.
DRP can use a Time Phased Order Point (TPOP) approach to plan orders at the branch warehouse level.
These orders are exploded via MRP logic to become gross requirements on the supplying source enabling the
translation of inventory plans into material flows. In the case of multi-level distribution networks, this explosion
process continues down through the various levels of regional warehouses, master warehouse, and factory
warehouse and finally becomes an input to the master production schedule.
ptg6843605
distributor − DMAIC
Page 109 The Encyclopedia of Operations Management
See distribution center (DC), Enterprise Resources Planning (ERP), logistics, Materials Requirements
Planning (MRP), warehouse, Warehouse Management System (WMS).
distributor – A wholesaler that buys, stores, transports, and sells goods to customers.
Distributors usually sell products made by others. However, it is common for distributors to conduct some
limited “value-adding” operations, such as cutting pipes or packaging potatoes. Although distributors are
considered customers for the manufacturer, they are not the end customer or consumer. It is wise for
manufacturers to hear the voice of the customer from the consumer’s perspective and not just the distributor’s.
See broker, channel conflict, cross-docking, distribution center (DC), distribution channel, inventory
management, logistics, supply chain management, wholesaler.
division of labor – Dividing a job into small, simple, standard steps and assigning one worker to each step.
Frederick Taylor (1911) promoted the concept of dividing work into small pieces so workers could quickly
learn jobs without much training. Division of labor and standardization of parts led to rifles made by several
people in the 1800s and the Model T Ford in the 1900s. Division of labor is the opposite of job enlargement, a
practice that has workers take on more tasks rather than fewer.
In the last thirty years or more, many managers have found that taking division of labor too far can lead to
boredom, does not develop the whole person, and does not build a learning organization. Division of labor also
creates many queues and waits and requires more coordination and supervision. Thus, many process
improvement projects enlarge jobs to remove queues and reduce cycle time.
On the other hand, some organizations report situations where processes can be improved by dedicating
individuals or teams to certain process steps. For example, Mercy Hospital in Minnesota found that having a
team of two people dedicated to the receiving process improved both quality and cost. Both vendor managed
inventories and outsourcing can be viewed as examples of division of labor, where the work is divided into
pieces that are done internally and other pieces that are done by other organizations. Division of labor is similar
to value chain analysis that evaluates outsourcing and insourcing alternatives.
See delegation, human resources, job design, job enlargement, scientific management, standardized work,
value chain.
diversion – The practice of selling products in unauthorized markets; also called parallel trade.
Gray (or grey) market resellers often acquire unwanted merchandise, overstocked products sold at a discount,
obsolete products sold at a discount, and products intended for another market (such as an international market)
and “divert” them to another market unintended by the manufacturer. In some cases, the practice is legal, but in
other cases resellers engage in theft, counterfeiting, diluting, and misrepresenting the products. This is an issue
for a wide variety of consumer products, such as health and beauty (hair products, cosmetics), pharmaceuticals,
consumer packaged goods, beverages, music, auto, and electronics. In many cases, service is not available or the
product warranty is invalid for gray market goods.
See gray market reseller.
DMADV – See Design for Six Sigma (DFSS), lean sigma.
DMAIC – A lean sigma problem-solving approach with five steps: Define, Measure, Analyze, Improve, and
Control.
Lean sigma projects are usually managed with a five-step problem-solving approach called DMAIC
(pronounced “Dee-MAY-ic”). These steps are described in the table below.
Define
Define the problem, scope, metrics, team, and sponsor.
Measure
Collect data and then measure and map the “as-is” state.
Analyze
Identify the system of causes and develop and test hypotheses (solutions, countermeasures). Select
the best set of solutions, including just do it, implement now, and implement later solutions.
Improve
Implement the best solutions.
Control
“Sustain the gains” by creating new roles and responsibilities, standard operating procedures
(SOPs), job descriptions, metrics, and reviews. Also, share learning and identify potential projects.
Source: Professor Arthur V. Hill
In many firms, the DMAIC process is “gated,” which means that the project team is not allowed to progress
to the next step until the master black belt or sponsor has signed off on the step. In the new product development
literature, this is called a phase review or stage-gate review. This author asserts that many DMAIC projects have
ptg6843605
dock – downtime
The Encyclopedia of Operations Management Page 110
too many gates that slow the project down. This author and many consultants now recommend having only two
gates for a process improvement project: a midterm report to ensure that the team is on track and a final report
when the project is complete (or nearly complete).
Most DMAIC projects find a number of “quick hits” early in the project. These should be implemented
immediately and should be documented so the project sponsor and the program champion can capture and
compare all project benefits and costs.
See deliverables, Design for Six Sigma (DFSS), Gauge R&R, kaizen workshop, lean sigma, PDCA (Plan-Do-
Check-Act), phase review, quick hit, sigma level, sponsor, stage-gate process.
dock – A door and platform used to receive and ship materials, usually from trailers; also called a loading dock or
receiving dock.
See cross-docking, dock-to-stock, logistics, trailer, warehouse.
dock-to-stock – The practice of moving receipts from the receiving dock directly to inventory without inspection.
Dock-to-stock eliminates the customer’s incoming inspection cost and reduces handling cost. However, it
requires that suppliers assure good quality products.
See dock, incoming inspection, logistics, point of use, receiving, supplier qualification and certification,
Warehouse Management System (WMS).
DOE – See Design of Experiments (DOE).
dollar unit sampling – An auditing technique for stratified sampling transactions that allows the auditor to make
statistically reliable statements about the misspecification error.
The auditing profession has considered the problem of how to make statistically reliable statements about an
audit when some transactions are more important than others and when the probability of a defect is very small.
The best approach is an extension of stratified random sampling called dollar unit sampling. In the auditing
literature, dollar unit sampling is also known as probability proportionate to size sampling and monetary unit
sampling. Roberts (1978, p. 125) stated, “When the proportion of population units with monetary differences is
expected to be small and the audit objective is to test for the possibility of a material overstatement, dollar unit
sampling is the best statistical technique.” A dollar unit sampling audit enables the auditor to make statistical
statements about the results of an audit, such as “Based on this audit, we are 95% confident that the total
overstatement amount for this population is no more than $500.”
See confidence interval, Poisson distribution, sample size calculation, sampling.
dot-com – An adjective used to describe companies that sell products or services over the Internet.
Dot-com companies usually do not sell products or services through “brick-and-mortar” channels. Products
are typically ordered over the Internet and shipped from warehouses directly to the customer. Services are
typically information services provided through the Internet. Amazon.com and eBay.com are two of the best-
known dot-com firms. Dot-com firms can be either business-to-business (B2B) or business-to-consumer (B2C)
firms. In some cases, dot-com firms can disintermediate traditional distribution firms.
See B2B, B2C, click-and-mortar, disintermediation, supply chain management.
double exponential smoothing – See exponential smoothing.
double marginalization – An economics term that describes a situation in which two firms in a supply chain have
monopoly power and each producer adds its own monopoly mark-up to the price.
The price of the finished product is higher than it would be if the two producers were vertically integrated.
See economics.
double sampling plan – See acceptance sampling.
downstream – See upstream.
downtime – Time that a resource (system, production line, or machine) is not working; called an outage in the
power generation context and a crash in the computer context.
Downtime is often separated into planned and unplanned downtime. Causes of planned downtime include:
Set-up – Adjustments required to prepare a resource to produce.
Start-up – The time from the end of set-up until the first good units are produced.
Cleaning – All activities required to remove materials and sanitize a process.
Changeover – Time to change from making the last unit of one product to the first good unit of the next.
ptg6843605
DPMO − DuPont Analysis
Page 111 The Encyclopedia of Operations Management
Operational downtime – Production stoppages imposed by the process for equipment and quality checks.
Maintenance – Scheduled (preventive) maintenance activities.
Personal Time – Line stoppage for meal breaks, shift changes, meetings, and personal time.
Construction – Downtime for re-layout or building construction.
Causes of unplanned downtime include machine failure (followed by emergency maintenance) and quality
problems that require the machine or line to stop.
The opposite of downtime is uptime. Service level agreements (SLAs) often specify guaranteed uptimes.
See capacity, operations performance metrics, Overall Equipment Effectiveness (OEE), reliability, Service
Level Agreement (SLA), setup time, Total Productive Maintenance (TPM).
DPMO – See Defects per Million Opportunities (DPMO).
DPPM – See Defective Parts per Million (DPPM).
drop ship – A logistics term for a shipment that goes directly from the supplier to the buyer without the seller
handling the product; often spelled dropship; sometimes called direct store shipment (DSD) or direct ship.
When a seller (typically a distributor or retailer) has a supplier (often a manufacturer or distributor) send a
shipment directly to a buyer (customer), the order is drop shipped to the customer. The seller does not handle
the product or put it in inventory. The supplier has little or no communication with the buyer. The buyer pays
the seller and the seller pays the supplier. Drop shipments reduce the customer’s leadtime and the seller’s
inventory handling cost, but usually increase the distribution cost because a small shipment (less than truck load)
must be made to the customer. The term “drop ship” is sometimes also used to describe a shipment to a different
location than the customer’s normal shipping location. Drop ship can be used as either a noun or a verb.
See distribution center (DC), logistics.
DRP – See Distribution Requirement Planning (DRP).
Drum-Buffer-Rope (DBR) – A Theory of Constraints (TOV) concept that sends a signal every time the
bottleneck completes one unit, giving upstream operations the authority to produce.
DBR is a production control system based on the TOC philosophy. Like other TOC concepts, DBR focuses
on maximizing the utilization of the bottleneck (the constrained resource) and subordinates all non-bottleneck
resources so they meet the needs of the bottleneck.
Drum – Completion of one unit at the bottleneck is the drum that signals (authorizes) all upstream
workcenters to produce one unit. The unconstrained resources must serve the constrained resource.
Buffer – A time cushion used to protect the bottleneck from running out of work (starving).
Rope – A tool to pull production from the non-bottleneck resources to the bottleneck. The DRB concept is
very similar to the “pacemaker workcenter” concept used in lean manufacturing.
See CONWIP, lean thinking, pacemaker, POLCA (Paired-cell Overlapping Loops of Cards with
Authorization), Theory of Constraints (TOC), upstream.
DSI – See The Decision Sciences Institute.
DSD – See drop ship.
DSM – See Design Structure Matrix.
dual source – The practice of using two suppliers for a single component; also called multiple sourcing.
Multiple sourcing is the use of two or more suppliers for the same component. Some people use the term
“dual source” to mean two or more suppliers. This is in contrast to sole sourcing, where only one supplier is
qualified and only one is used, and single sourcing, where multiple suppliers are qualified, but only one is used.
See single source.
due diligence – A careful evaluation done before a business transaction.
A common example of due diligence is the process that a potential buyer uses to evaluate a target company
for acquisition. Wikipedia offers a very thorough discussion of this subject.
See acquisition, purchasing.
ptg6843605
dunnage – DuPont STOP
The Encyclopedia of Operations Management Page 112
dunnage – Fill material used to minimize movement within a container to protect products being shipped.
Dunnage can be any packing material, such as low grade lumber, scrap wood, planks, paper, cardboard,
blocks, metal, plastic bracing, airbags, air pillows, bubble wrap, foam, or packing peanuts, used to support,
protect, and secure cargo for shipping and handling. Dunnage can also be used to provide ventilation and
provide space for the tines of a forklift truck.
See logistics.
DuPont Analysis – An economic analysis that can be used to show the return on investment as a function of
inventory and other economic variables.
Operations managers can use the DuPont Analysis to analyze the impact of changes in inventory investment
on Return on Investment (ROI). The DuPont Analysis can be used to show how much (1) the carrying cost goes
down when inventory goes down, (2) profit (return) goes up when the cost goes down, (3) investment goes down
when inventory investment goes down, and finally, (4) ROI goes up dramatically as the numerator (return) goes
up while the denominator (investment) goes down at the same time. This assumes that revenue is not affected by
inventory, which may not be true for a make to stock firm unless the inventory reduction is managed very
carefully. From an inventory management point of view, the DuPont Analysis is less important when interest
rates and carrying charges are low.
A DuPont Analysis shows the sensitivity of the firm’s ROI to changes in input variables (drivers), such as
inventory. (Note: Many organizations change ROI to an EVA or economic profit calculation.) This analysis is
similar to a strategy map and a Y-tree because it shows the drivers of higher-level performance metrics.
The following is an example of a DuPont Analysis for a hypothetical firm.
See balanced scorecard, financial performance metrics, inventory turnover, Return on Net Assets (RONA),
strategy map, what-if analysis, Y-tree.
DuPont STOP – A safety training program developed at DuPont.
The Safety Training Observation Program (STOP) is a widely used safety program that teaches workplace
safety auditing skills, with steps to reinforce safe work practices and correct unsafe practices. STOP is based on
the following principles:
Labor Selling expense Sales
$800 $150 $4,250
Materials Cost of goods sold Operating cost Operating profit
$950 $3,125 $3,833 $225
Overhead General expense Tax
$1,375 $210 $192
Inventory carrying charge Inventory carrying cost
25.00% $348
Return on net
assets (RONA)
Cash
5.76%
$1,000
Raw Material Receivables Current assets
$325 $820 $3,210
Margin
5.30%
Work in process Inventory
$500 $1,390 Net assets *
Inventory turnover
$3,915 2.25
Finished goods Property/Plant * Net assets is identical to invested capital.
$565 $128
Equipment Fixed assets
$425 $705
Other assets
$152
Source: Professor Hill’s Excel workbook DuPont.xls
ptg6843605
durability − Early Supplier Involvement (ESI)
The Encyclopedia of Operations Management Page 113
All injuries can be prevented.
Employee involvement is essential.
Management is responsible for safety.
All operating exposures can be safeguarded.
Safety training for workers is essential.
Working safely is a condition of employment.
Management audits are a must.
All deficiencies must be corrected promptly.
Off-the-job safety will be emphasized.
See Occupational Safety and Health Administration (OSHA), safety.
durability – A quality term used to refer to a product’s capability to withstand stress, wear, decay, and force
without requiring maintenance; also called durable.
See durable goods, quality management, robust, service guarantee.
durable goods – Products that people keep for a long time and have useful lives of more than five years; also
known as capital goods or hard goods.
Examples of durable products include cars, furniture, and houses. Durable goods can usually be rented as
well as purchased. In contrast, non-durable goods are almost never rented.
See consumable goods, durability, Fast Moving Consumer Goods (FMCG), white goods.
Durbin-Watson Statistic – A statistical test for first-order autocorrelation (serial correlation) in time series data.
Autocorrelation is the correlation between a variable in one period and the previous period. For example, the
weather temperature is highly autocorrelated, which means that the correlation between the weather in one day
tends to vary with the weather on the previous day. If today is hot, then so is tomorrow.
The Durbin-Watson Statistic is used to test for first-order autocorrelation in time series data. It is most
commonly used to test for autocorrelation in the residuals for regression models that deal with time series data. It
is often also used to test for autocorrelation in the forecast error for a time series forecasting model. The term
“error” (e
t
) is used to mean either the residuals from a regression or the forecast error for a forecasting model.
The Durbin-Watson test compares the error in period t with the error in period t−1. The following equation
is the Durbin-Watson test statistic
2 2
1
2 1
( )
n n
t t t
t t
d e e e
, where t is the time period, e
t
is the residual in
period t, and n is the total number of observations available. The statistic (d) is constrained to the range (0, 4)
with a midpoint of 2. A value of d close to 2 suggests that the time series has no autocorrelation. A low value of
d (close to zero) implies positive autocorrelation because the differences between e
t
and e
t−1
are relatively small.
A high value of d (close to four) implies negative autocorrelation because the differences between e
t
and e
t−1
are
relatively large.
The Durbin-Watson test can be used to test for both positive and negative autocorrelation. However, the null
hypothesis is usually no significant autocorrelation and the alterative hypothesis is positive autocorrelation.
Tables for the Durbin-Watson test can be found in many standard statistics texts, such as Kutner, Neter,
Nachtsheim, and Wasserman (2004).
See autocorrelation, Box-Jenkins forecasting, linear regression, runs test, time series forecasting.
Dutch auction – An auction method where the price is lowered until a bidder is prepared to pay; also known as a
descending bid auction.
In a Dutch auction, the auctioneer begins with the seller’s asking price and then lowers the price until a
bidder is willing to accept the price or until a predetermined reserve price (the seller’s minimum acceptable
price) is reached. The winning bidder pays the last announced price. The Dutch auction is named for the Dutch
tulip auctions in the Netherlands.
See e-auction, e-business, e-procurement, reverse auction, sniping.
E
EAN (European Article Number) – See Universal Product Code (UPC).
earliness – See service level.
ptg6843605
early detection – e-auction
The Encyclopedia of Operations Management Page 114
early detection – The quality management concept that it is better to find and fix defects early in a process.
See catchball, cost of quality, defect, Fagan Defect-Free Process, mixed model assembly, quality at the
source, scrum, setup time reduction methods.
Early Supplier Involvement (ESI) – The collaborative product development practice of getting suppliers
involved early in the product design process.
Good suppliers have core competences around their product technologies. Therefore, firms that involve their
suppliers in product development at an early stage can take advantage of these core competencies and potentially
reap financial and competitive rewards.
Companies that involve suppliers early report the following benefits: (a) reduced product development time,
(b) improved quality and features, (c) reduced product or service costs, and (d) reduced design changes.
Companies do best when they give suppliers the leeway to come up with their own designs rather than simply
manufacturing parts to their customers’ specifications. Suppliers often have more expertise than their customers
in their product technologies.
See concurrent engineering, JIT II, New Product Development (NPD).
earned hours – The labor or machine hours calculated by multiplying the actual number of units produced during a
period by the standard hourly rate.
Efficiency is calculated as the ratio (earned hours)/(actual hours) during a period. For example, a workcenter
has a standard rate of one unit per hour. It worked eight actual hours today, but earned ten hours (i.e., it
produced then units). Therefore, the efficiency of this workcenter is (earned hours)/(actual hours) = 10/8 =
125%. Earned hours is similar to the practice of earned value management in project management.
See Earned Value Management (EVM), efficiency, productivity.
Earned Value Management (EVM) – A methodology used to measure and communicate the progress of a
project by taking into account the work completed to date, the time taken to date, and the costs incurred to date.
Earned Value Management (EVM) helps evaluate and control task/project risk by measuring progress in
monetary terms. EVM is sometimes required for commercial and government contracts. Under EVM, work is
planned, budgeted, and scheduled in time phased “planned value” increments, constituting a cost and a schedule
measurement baseline.
The description below applies EVM to a task; however, the same concept can easily be extended to an entire
project. Time and material is spent in completing a task. If managed well, the task will be completed with time
to spare and with no wasted materials or cost. If managed poorly, the task will take longer and waste materials.
By taking a snap-shot of the task and calculating the Earned Value, it is possible to compare the planned cost and
schedule with the actual cost and schedule and assess the progress of the task. When considering an entire
project, it is possible to extrapolate the schedule and cost to estimate the probable completion date and total cost.
The basics of EVM can best be shown on an S-curve. In its simplest form, the S-curve is a graph showing
how the task budget is planned to be spent over time. The three curves on the graph represent:
Budgeted cost for work scheduled – The budgets for all activities planned.
Actual cost of work performed – The actual costs of the work charged so far.
Budgeted cost of work performed – The planned costs of the work allocated to the completed activities.
Earned Value is defined as the percentage of the project complete times the project budget. The schedule
variance is the difference between the Earned Value and the budget. Cost variance is the difference between the
Earned Value and the actual costs of the works.
The benefits for project managers of the Earned Value approach come from:
Disciplined planning using established methods.
Availability of metrics that show variances from the plan to generate necessary corrective actions.
See critical chain, earned hours, project management, work breakdown structure (WBS).
e-auction – A Web-based tool for making a market more efficient.
The best example of an electronic auction is the popular ebay.com. A reverse auction is where the buyer
calls for bids for something from potential suppliers. For example, General Electric will notify a group of
qualified suppliers that they are invited to participate in an electronic auction. The date and product
specifications are defined by the buyer. At the time of the auction, the participating bidders assemble at a
common Internet site and bid for the contract.
ptg6843605
EBITDA − Economic Order Quantity (EOQ)
Page 115 The Encyclopedia of Operations Management
See Dutch auction, e-business, e-procurement, reverse auction, sniping.
EBITDA – Earnings Before Interest, Taxes, Depreciation, and Amortization; an indicator of a company’s financial
performance calculated as revenue minus expenses (excluding tax, interest, depreciation, and amortization);
sometimes called EBIDTA (Earnings Before Interest, Depreciation, Taxes, and Amortization).
EBITDA is an approximate measure of a company’s operating cash flow based on data from the company’s
income statement. EBITDA is calculated by looking at earnings before the deduction of interest expenses, taxes,
depreciation, and amortization. This measure of earnings is of particular interest in cases where companies have
large amounts of fixed assets that are subject to heavy depreciation charges (such as manufacturing companies).
Because the accounting and financing effects on company earnings do not factor into EBITDA, it is a good way
to compare companies within and across industries. This measure is also of interest to a company’s creditors,
because EBITDA is essentially the income that a company has free for interest payments. In general, EBITDA is
a useful measure only for large companies with significant assets or a significant amount of debt financing. It is
rarely a useful measure for evaluating a small company with no significant loans.
EBITDA is a good metric to evaluate profitability, but not cash flow. Unfortunately, however, EBITDA is
often used as a measure of cash flow, which is a very dangerous and misleading thing to do because there is a
significant difference between the two. Operating cash flow is a better measure of how much cash a company is
generating because it adds non-cash charges (depreciation and amortization) back to net income and includes the
changes in working capital that also use/provide cash (such as changes in receivables, payables, and inventories).
These working capital factors are the key to determining how much cash a company is generating. If investors
do not include changes in working capital in their analyses and rely solely on EBITDA, they may miss clues that
indicate whether a company is losing money because it cannot sell its products.
See financial performance metrics, income statement.
e-business – The conduct of business over the Internet; sometimes used interchangeably with e-commerce.
Bartels (2000) defines e-commerce as using electronic network support for customer- and supplier-facing
processes and e-business as using electronic network support for the entire business. Based on these definitions,
he argues that e-business is more difficult that e-commerce because it involves integration e-commerce activities
with back office operations such as production, inventory, and product development.
See B2B, B2C, back office, Dutch auction, e-auction, e-procurement, extranet, intranet, reverse auction,
sniping.
ECO – See Engineering Change Order (ECO).
e-commerce – See e-business.
econometric forecasting – A forecasting method that considers a number of different leading economic indicators,
such as disposable income and meals away from home, to make forecasts; also known as extrinsic forecasting.
Econometric models use leading indicators to make forecasts. For example, a sharp rise in the cost of
gasoline may well be a good indicator (predictor) of an increased demand rate for fuel-efficient cars.
Most econometric studies use multiple regression models. For example, the Onan Division of Cummings
Engine developed a regression model and found that fast-food sales and disposable income could be used to
forecast the sales of recreational vehicles one quarter ahead.
See Box-Jenkins forecasting, forecasting, leading indicator, linear regression.
Economic Lot Scheduling Problem (ELSP) – A class of lotsizing problems that involves finding the optimal
(or near optimal) order size (or cycle length) to minimize the sum of the carrying and ordering (setup) costs for
multiple items that share the same capacity resource (the “bottleneck”).
Even though the problem has the word “scheduling” in its name, it is a lotsizing problem rather than a
scheduling problem. This ELSP methodology can be implemented fairly easily for any manufacturing process
that builds to stock and has a fairly level demand throughout the year. The procedure finds the optimal order
intervals
*
i
n
(periods supply) for each item i. When the demand varies over time, it is better to use the periods
supply than the order quantities. It may also be important to find the optimal safety stock for these order
quantities, where the leadtime for each item is based on the order interval. The time phased order point system
can then be used to determine order timing.
See algorithm, Economic Order Quantity (EOQ), lotsizing methods, run time, setup time.