Nof S.Y. Springer Handbook of Automation

Подождите немного. Документ загружается.

1265

Automating I

71. Automating Information and Technology Services

Parasuram Balasubramanian

While the information services industry dates

back to the 15th century AD, the information

technology services is as young as 60 years. Yet

the pace at which both have grown is phe-

nomenal, and this is tied to the evolution

of computer technology. Networking of com-

puters and people through the Internet has

moved into high gear. It has also given birth

to multiple business segments in the pro-

cess. Delivery of data and information, data

processing, business process outsourcing, an-

alytics, and printing and display solutions are

the ﬁve segments identiﬁed within the in-

formation services. Computer-aided software

engineering, independent software testing

and quality assurance, package and bespoke

software implementation and maintenance,

network and security management, and host-

ing and infrastructure management are covered

as segments within information technology

services. The automation path of each seg-

ment is reviewed in detail. An impact analysis

to identify the changing landscape is de-

scribed. Finally current trends are traced,

followed by predictions for future develop-

ments.

71.1 Preamble ............................................ 1265

71.1.1 Evolution of the Information

and Technology Services Industry .. 1266

71.1.2 The Opportunity for Automation.... 1266

71.2 Distinct Business Segments .................. 1267

71.3 Automation Path

in Each Business Segment .................... 1269

71.3.1 Delivery of Data and Information .. 1269

71.3.2 Data Processing........................... 1269

71.3.3 Business Process Outsourcing........ 1271

71.3.4 Analytics .................................... 1272

71.3.5 Printing and Display Solutions ...... 1273

71.4 Information Technology Services........... 1274

71.4.1 Computer-Aided

Software Engineering .................. 1274

71.4.2 Independent Software Testing

and Quality Assurance ................. 1275

71.4.3 Package and Bespoke Software

Implementation and Maintenance 1277

71.4.4 Network

and Security Management............ 1278

71.4.5 Hosting

and Infrastructure Management ... 1280

71.5 Impact Analysis ................................... 1281

71.6 Emerging Trends ................................. 1282

References ................................................. 1282

71.1 Preamble

The design speciﬁcation for the electronic voting ma-

chines (EVMs) used in 2004 national elections in India

had to take account of the high level of illiteracy among

citizens, cater for more than 14 different languages,

tackle poor infrastructure in terms of road network and

power availability in rural polling stations, and accept

manual methods of voter authentication [71.1]. Nearly

1.3 million EVMs adhering to the above speciﬁcation

were deployed in the election, in which 330 million cit-

izens voted. The election process went off without any

signiﬁcant problem with respect to the EVMs. Mean-

while inEurope, near Paris, a high-technology company

has been engaged in using crash simulation software on

Intel personal computers (PCs) to model and test new-

generation vehicles to meet regulatory requirements

without building expensive prototypes [71.2]. Not to be

left behind in the innovation race, many ﬁrms in Europe

and the USA have been busy delivering portable car

navigation systems using the global positioning system

(GPS) to the mass market.

Part G 71

1266 Part G Infrastructure and Service Automation

These three cases represent the current state of the

art in automation in information and technology ser-

vices industries. The contexts are different and the level

of automation varies, yet all three are powerful exam-

ples of information technology making its impact on

mankind. The transformational power of this industry,

as it impacted on the USA from the colonial period to

current times, is traced by Chandler andCortada [71.3].

The progress achieved is signiﬁcant and exponen-

tial over the past 60years, ever since computers entered

this arena. As a general-purpose machine, the computer

has invaded almost all aspects of human life. Comput-

erization has resulted in productivity gains that could

not have been dreamt of until the 20th century. It has

been a process of unceasing innovations, integration

of diverse devices, seamless merger of hardware and

software, and evolution of new market paradigms. It

has also given birth to an industry focused on provid-

ing information technology (IT) services to companies,

governments, universities, and the public at large. It is

worth studying this process to understand the growth of

this industry and where it is headed.

71.1.1 Evolution of the Information

and Technology Services Industry

Johann Gutenberg introduced the printing press in

1436 AD in Europe and is credited for opening the

doors to the Renaissance and the scientiﬁc revolu-

tion [71.4]. The information services industry was born

with the mass production of books (within the next

50years 8 million books were printed) and it facilitated

sharing of knowledge among scholars, particularly in

Europe [71.5, 6]. The mass media industry consisting

of newspapers, magazines, radio, television, and adver-

tising, however, had to wait until the 19th century (for

Marconi, Edison, and Walter Thompson to give birth

to). A brief view of its growth in the USA is con-

tained in [71.7]. The invention of the telegraph and

Morse code by Morse in 1835 and the telephone in

1870 by Graham Bell, and their rapid spread of us-

age, along with the above, ensured that the spoken word

joined the printed one to be shared across the world.

Emile Berliner, standing on the shoulders of Edison

and other eminent scientists, patented the gramophone,

thus paving way for the music industry. A comprehen-

sive view of each of these inventions and inventors is

available on the Internet [71.8].

However, a parallel development over the centuries,

in mathematics coupled with mechanical and electronic

devices, led the development of information technol-

ogy. Calculators (Pascal, 1642) gave way to the analytic

engine of Babbage (1832), and Hollerith’s tabulating

machine was used ﬁrst in the 1890 US census. The mid-

dle of the 20th century saw the invention of computers

and the world has been spinning fast ever since. Along

with computers, the informationtechnology services in-

dustry was born in the 1950s and has grown in multiple

dimensions [71.9].

Computer architecture separated the data input from

its processing and output layers and exploited the dis-

tinction between data and logic. Advances in input

devices, storage devices, processors, and data presen-

tation layer during the past 60years have come about

by the marriage of information services technology

components with computer devices. The distinction

between telephone, television, and computer is blur-

ring and feeding the growth of services at a furious

pace.

71.1.2 The Opportunity for Automation

The ﬁrst generation of computers could perform various

mathematical and logical operations while the data had

to be stored in an external media. The machine had very

little capacity to hold data. The growth in computeriza-

tion over the next six decades can be studied in multiple

dimensions. The technological developments in the pro-

cessor, input devices, and output devices deﬁne the ﬁrst

dimension, while advances in software technology con-

stitute the second. Along the way there was integration

among these devices as well as embedding of software

into hardware, which played a catalyst’s role in further

automation.

Computers werenetworked during the 1970s,which

paved way for the next wave of automation of tasks

carried out in multiple locations by diverse groups.

They acquired online and real-time capabilities, thus

forsaking their batch processing culture. It did not take

long for human ingenuity to invent the multiprocess-

ing, multitasking systems, network security, and virtual

authentication solutions. The powerful combination of

hardware and software with these features resulted in

many more activities becoming automated.

If computers are capable of storing vast amounts of

data and perform complex processes in seconds, can

they then be asked to perform the role of experts (ex-

pert systems), acquire human-like intelligence (called

artiﬁcial intelligence) or at the least detect patterns and

extract knowledge out of available data (knowledge-

based systems)? Commonly grouped under the title of

machine learning, these topics have become the en-

Part G 71.1

Automating Information and Technology Services 71.2 Distinct Business Segments 1267

during quest of many researchers over the past three

decades.

Work ﬂow is integral to all processes performed

manually or otherwise. Work ﬂow automation became

a logical target for development in the early 1980s,

coupled with the development of groupware that facili-

tated concurrent working of multiple teams. Computers

had stepped out of the data processing and data stor-

age zone by this time. They were capable facilitators

for information sharing and multigroup multiprocess

working.

Introduction of personal computer, starting from

the late 1970s, into the global market brought a fresh

paradigm to computing and automation. The gates

were opened for mass application and coverage of

every human endeavor by PCs. Enormous creativity

was unleashed by technophiles all over the world.

It was no longer conﬁned to data processing; text

and word processing became common. Slide rules

and calculators gave way to spreadsheets. Transparen-

cies were no longer the preferred means of making

presentations as they were replaced by PowerPoint

slides.

Impressive as they are, these developments pale in

signiﬁcance when the impact of the Internet is consid-

ered. Since the invention of the World Wide Web and

the browser technology by Tim Berners Lee and Marc

Andreessen, the universe of computers has been ﬂipped

over onceagain. Inless than a decade ofits introduction,

the Internet phenomenon has lead Thomas Friedman to

declare that it is a ﬂat world.

The power of the Internet has been fully lever-

aged by electronic mail from the mid 1990s. Hotmail,

founded by Sabhir Bhatia, became an instant hit as it

promised and provided connectivity from anywhere to

anyone. The data and text sharing power of comput-

ers and the Internet has been enhanced to recognize

voice, feel (through touch screens), still and video im-

ages, and to recognize patterns. Convergence of devices

has accelerated the use of the computer as a phone

and a television too. Apple made a signiﬁcant business

breakthrough with its iPod to share music through the

Internet.

Meanwhile the business community has enabled

internet-based transactions, including ﬁnancial. Univer-

sities have enriched their student interaction and grown

the online and distance education segments. And the

government sector has discovered the potential of this

technology by facilitating digital submission of tax re-

turns and other ﬁlings. In short, this revolution has

touched every segment of human activity and innova-

tion, and automation at unprecedented levels continues

to fuel its breadth and depth.

These developments have impacted on every busi-

ness segment of the information and technology ser-

vices, which is the focus of our study. We will identify

the key segments and study each in depth in the follow-

ing Sections.

71.2 Distinct Business Segments

The information and technology services industry is not

a homogenous industry but rather consists of many seg-

ments, each with distinct characteristics. At a macro

level, information services (Sect.71.3) need to be sepa-

rated from information technology services (Sect.71.4).

Five signiﬁcant segments can be recognized within

the information services (Sect.71.3), consisting of de-

livery of data and information (Sect.71.3.1), data

processing (Sect.71.3.2), business process outsourcing

(Sect.71.3.3), analytics (Sect.71.3.4), and printing and

display solutions (Sect.71.3.5).

Media such as newspapers, journals and magazines,

radio and television, and the Internet serve as chan-

nels of information delivery. The content delivered can

vary from political and social news events to weather

and sports or stock and commodity prices. Reuters and

Bloomberg are examples of a speciﬁc type of content

providers. Automation-induced changes in this industry

can be tracked by studying these two ﬁrms over the past

decades (Sect.71.3.1).

Data processing as a service commenced in the

mainframe era when computers were expensive and re-

quired special skills to program and operate. It was

typical to do business functions such as ﬁnancial ac-

counting or payroll processing or reconciliation of

interbank transactions or process examination results

in large universities using third-party data process-

ing services (Sect.71.3.2). Technological advances in

computers and allied peripherals have impacted heav-

ily on this segment; hence it is worthy of separate

classiﬁcation.

Companies the world over have found the business

environment becoming complex due to geographical

spread, globalization, product variety, and legislation

Part G 71.2

1268 Part G Infrastructure and Service Automation

to be complied with. Many times these challenges

have resulted in attention and key resources being di-

verted away from core business to auxiliary purposes.

Hence, companies have longed for service providers

with appropriate expertise and scale to whom they

could outsource auxiliary business processes such as

employee beneﬁts management and insurance claims

processing. Networking of computers, proliferation of

desktops, and the ubiquityof theInternet duringthe past

two decades have enabled the emergence of third par-

ties who can provide these business process outsourcing

(BPO) services (Sect.71.3.3). While outsourcing data

processing results in only the data being handed over,

business processing leads to data, (occasionally) peo-

ple, and processes being handed over too. The service

provider then hasgreat opportunity to innovateand sim-

plify the processes on an ongoing basis.

Data and business processing services are strictly

transaction processing solutions that handle enormous

volume of data. They are however governed by the poli-

cies and decisions taken by the ﬁrm; for example, an

airline with a goal to maximize revenue would decide

on creating multiple classes of service in a given ﬂight,

price them differentially, and divide the total seats avail-

able into these classes based on a rationale. This is

called revenue management. Apart from sellingseats on

an ongoing basis, the airline has to assess and evaluate

the effectiveness of these decisions periodically. Ana-

lytical processing is a line of service facilitating such

reviews and revisions (Sect.71.3.4). The skill sets re-

quired and the tools utilized in analytical processing are

very different from those relating to data or business

processing.

Advances in technology have changed the way

in which information is delivered as an output

(Sect.71.3.5). Convergence of telephone, television,

and computer technologies has created innovative so-

lutions that were hard to imagine a few decades ago.

Automation of mechanical, electrical, and electronic

devices when coupled with computers has opened up

a new domain for control systems; these developments

are to be studied in this segment.

The information technology services (Sect.71.4)

can be segmented into computer-aided software en-

gineering (Sect.71.4.1), independent software test-

ing and quality assurance (Sect.71.4.2), package

and bespoke software implementation and mainte-

nance (Sect.71.4.3), network and security management

(Sect.71.4.4), and hosting and infrastructure manage-

ment (Sect. 71.4.5). While these are by no means

exhaustive, they represent the key segments at present.

Software development has progressed rapidly from

machine-level coding to programmingin high-level lan-

guages to an environment with almost no need to code.

The developer community recognized the need and fea-

sibility of automating many of the tasks they perform,

early in the game. Computer-aided software engineer-

ing (CASE) tools emerged when their generic utility

was discovered (Sect.71.4.1). This segment is to be

explored in detail.

As the size of software code has increased, so has

the time and effort required to test it. Nearly 30% of

the total effort for developing software is consumed by

the testing activity alone. Testing is both manpower and

domain knowledge intensive. It calls for comprehen-

sive understanding of the business environment within

which the software has to function, the assembly of

platforms and tools used, and the methodology used

for product development. While many software bugs

can be innocuous, some can have a disastrous impact

on system deliverables. Automation has changed the

way in which testing and quality assurance tasks are

performed in many organizations. Third-party vendors

have emerged to provide this service independent of the

developers (Sect.71.4.2).

The landscape of systems has undergone a meta-

morphosis from its early days to current periods. Every

system was a custom-developed system (known as be-

spoke) to start with. However the opportunity to reuse

design and code was spotted soon. The other end of this

spectrum consists of packaged software which has the

highest level of efﬁciency in terms of code reuse, be-

sides the obvious beneﬁt of quick implementation. In

large corporations it is common for both to coexist; for

example, a bank may have core banking functions im-

plemented through packaged software but its ﬁnancial

accounting and reporting may still be in a custom-

developed system running for decades. The year-2000

(Y2K) scenario is the well-understood instance here.

Once a system is developed, implementing and main-

taining it is time, cost, and manpower consuming. The

role played by automation in this segment (Sect.71.4.3)

is a fascinating study.

Proliferation of computers and associated software

has given rise to newer management issues and chal-

lenges. When thousands of desktops are strung together

with dozens of mini and mainframe computers and data

moves across many departments and geographical lo-

cations, failure in any part of the network is bound to

adversely affect the network as a whole. Besides, net-

works have to be managed for efﬁciency of resource

usage as well. An allied concern is protection of data

Part G 71.2

Automating Information and Technology Services 71.3 Automation Path in Each Business Segment 1269

from unauthorized access. Network and security man-

agement (Sect.71.4.4) has emerged as a major segment

of information technology that has attracted ﬁrms with

appropriate expertise and willingness to absorb associ-

ated risks. Product vendors have also emerged in this

space, offering products to protect against viruses, in-

trusions with intent to steal data, etc.

Computers and information technology have be-

come the backbone of many enterprises, commercial or

otherwise. Yet they are not the core functions of tra-

ditional manufacturing ﬁrms or service providers such

as banks, stock markets, and insurance companies. As

computerization has matured over the decades, many

ﬁrms have also aspired to hand over the entire set

of activities and resources to expert service providers.

Alternatively, vendors invest in such resources and pro-

vide hosted services at an appropriate charge. Hosting

and infrastructure management services (Sect.71.4.5)is

thus a distinct segment in this domain.

The ten segments identiﬁed within the information

and technology industry are studied in detail in the fol-

lowing sections. They are by no means exhaustive and

they constitute the most signiﬁcant segments at present,

tet the scenario may be vastly different in a decade from

now due to the rapid changes and developments taking

place in this industry.

71.3 Automation Path in Each Business Segment

71.3.1 Delivery of Data and Information

Paul Reuters started a news and stock information

service from London’s ﬁnancial district in 1851 us-

ing his telegraphic skills and 200 pigeons. His ﬁrm

expanded its services rapidly to Asia and Amer-

ica within the following 20years. By 1923, Reuters

had started using radio transmission of data of stock

prices, exchange rates, and news. It diversiﬁed into

television (TV) media by the 1950s, and computer

screen display of stock and commodity prices in the

1970s. It joined hands with Merrill Lynch in 2005

to ﬂoat workstations for online ﬁnancial informa-

tion [71.10].

Michael Bloomberg founded his ﬁrm in 1982 to

provide online ﬁnancial data about companies and

their performance with dedicated terminals to market

analysts. By 1987 the ﬁrm diversiﬁed into launch-

ing a trading system and discontinued printing the

user manual soon. The printed document was replaced

by online features, online manual, and help screens.

Bloomberg expanded into travel, news services, pro-

fessional services, and email and got into magazines,

radio stations, TV, multimedia, website, online trad-

ing, books, foreign exchange trading and so on in the

ensuing decade [71.11].

Widespread use of the World Wide Web, during the

past decade, has created a mass market for information

services and has ensured that both the generation and

distributionservices are global and not dependent on the

size of the ﬁrm. Every corporation, big or small, proﬁt

or nonproﬁt, business or social entity has launched

its own website to share information. Thousands of

ﬁrms have jumped into this business of designing and

operating third-party sites. Very specialized or niche in-

formation, segmented by geography, age group, hobby

or special interest, alumni groups, are all available to-

day. Portals that serve as gateways have multiplied,

along with content service providers. The industry has

gained another orbit with the introduction of fast and

efﬁcient search engines from Yahoo and Google. Along

with the growth of this industry, issues such as security

of information, privacy protection of providers, and au-

thentication of sourcehave becomecritical. Theseneeds

in turnare serviced by another set of ﬁrmswith software

tools.

Expansion of services and diversiﬁcation of chan-

nels are facilitated by advances inthe computer industry

and the media convergence phenomena. The cable ser-

vice provider has become the Internet service provider

as well. The Internet is being leveraged by ﬁrms such

as Skype to facilitate teletalk and to view video-quality

images. Convergence between television, telephone,

and Internet devices is accelerating. Innovation man-

agement and product and market strategy issues are

coming to the fore, along with this technological con-

vergence [71.12]. Cellphones and other mobile devices

can carry voice, text, image, and data, and link to the

Internet. A fascinating future awaits us, with almost ev-

ery household or ofﬁce device becoming intelligent,to

transmit, store or accept, and process data.

71.3.2 Data Processing

Data processing services consist of data preparation and

data processing stages followed by data presentation to

Part G 71.3

1270 Part G Infrastructure and Service Automation

the user community. A comprehensive history of devel-

opments can be found in [71.13,14]

Data preparation activities have gone through four

major phases of evolution (Fig.71.1).

Phase 1 corresponds to the early years when data

was prepared in batch mode, ofﬂine, using data en-

try machines and punched cards. Edit and veriﬁcation

runs were performed on the computer to catch errors

in punched cards. The prime objective was to detect

and rectify deviations if any from the data contained

in the original documents. After this clean up a data

validation run was performed to identify logical errors

in the data presented. Range checks and ﬁeld compar-

isons were done before accepting a set of data for the

processing run. The sequential nature of these activities

and their time-consuming nature is apparent. During the

next decade, technological advancement from punched

cards to ﬂoppy drives (8



to 5.25



to 3.5



) contributed

to data storage efﬁciency and cost effectiveness. Par-

tial rectiﬁcation and validation could be performed in

the data entry system itself. New category of workers

called data entry operators and input/output clerks were

employed in the industry.

Once dumb terminals could be attached to comput-

ers and dispersed in a location, the online data capture

phase (phase 2) began, in the 1980s. This lead to elim-

ination of duplicated effort for data capture in many

instances. Data edit and validation functions could also

be performed online, thus saving effort, compressing

time, and reducing errors.

Phase 3 can be related to the period of introduc-

tion of PCs and networking them with the processing

Phase 1

Batch mode,

offline

Data entry, edit

and validation

Phase 2

Online

Data entry, edit

and validation

Phase 3

At source

Data capture

and real-time

validation

Phase 4

At source,

interactive

Data capture and

error prevention

Fig. 71.1 Data preparation phases

systems. At-source data capture became feasible with

real-time validation. Online help features started ap-

pearing and the next logical step was to prevent capture

of erroneous data. Graphical user interface, pull-down

menus, predeﬁned ﬁelds, and automated cursor move-

ments to control the sequence, all being software

advancements, played a key role in signiﬁcant enhance-

ment to data capture efﬁciency and effectiveness. This

was the period when job categories such as data entry

operators and input/output clerks disappeared from the

market place.

Phase 4, which is the current phase, is characterized

by extensive networking using the Internet, intranets,

and extranets. Data is captured not onlythrough PCsbut

through other interfacedevices,such astelephones, card

swiping machines, point-of-sale (POS) systems, and ra-

diofrequency identiﬁcation (RFID) readers etc., as well.

Captured information is not just data but can be voice

or image too. At-source data capture is now ubiquitous

since the device can be in a customer location, remote,

and man or machine; it is interactive for information

sharing or data capture and facilitates decision choice

for the user.

What does the future hold? Cross-validation of data

from disparatesystems (ifhe is buying anexpensive car,

has he been paying income taxes regularly?) is com-

ing. Person authentication through biometric sensors or

other means before accepting data is already present in

some systems but is likely to become universal. Fur-

ther smart systems are on the anvil for the future. The

intent is to make every device, be it a refrigerator, mi-

crowave ovenor armyuniform, sosmart that it transmits

information proactively and in an automated manner.

The data processing stage has evolved from manual

mounting of tape and disc drives, a human computer

operator ﬁring the sequence of operations through

a console, inserting stationery into the printer, and

managing it to total automation in ﬁve decades. The

progress here has been continuous and can be credited

to both hardware and software developments.

The operating system (OS) of the computer has

learnt to be multitasking and to sense device changes

automatically. It can act as per location or chronologi-

cal requirements. It can time-stamp transactions. It can

distribute work to multipleprocessors and assemble and

present the output in a seamless manner. It can protect

itself from intrusions and threats. It can also compile in-

formation regarding the system performance and create

an alarm when needed.

Data management has another aspect to it, namely

the database architecture. Developments in database

Part G 71.3

Automating Information and Technology Services 71.3 Automation Path in Each Business Segment 1271

management have progressed from simple and serial

records to hierarchical databases (evolved to opti-

mize data management during the tape drive era) to

network databases to relational databases (RDBM).

Codd [71.15]enunciated the basicprinciples of RDBMs

in the 1970s and thus facilitated the optimal way of

data storage and retrieval for all media, hierarchical or

random. Major advances in data processing have come

from integration of RDBMs with OSsaswellasstorage

media. Storage area networks (SAN) and network as

storage (NAS) concepts have evolved in the past decade

to access widely distributed storage devices and share

their utilization effectively.

71.3.3 Business Process Outsourcing

Businesses run due to interaction with customers, ven-

dors, and other stakeholders and based on the ﬂow of

transactions; for example, customers place orders with

the ﬁrm (indent), and the ﬁrm, upon delivery of goods,

seeks for the customer to pay (invoice). The payment

is effected through a banking transaction [check, draft,

electronic ﬁnancialtransfer (EFT)]. Note that goods and

funds ﬂow in the system but their ﬂow is controlled

through the ﬂow of information. Before an order is exe-

cuted, it is more than likely that a set of activities called

order processing happenswithin theﬁrm (sometimes in-

teracting with the customer). There can be many tasks

performed in order processing adhering to certain se-

quence and operating rules of the company. Most of the

time these tasks are performed by human beings.

As discussed earlier, it is cost effective for many

ﬁrms to outsource noncore business processes to

a third party and to derive operational and cost efﬁ-

ciency. Travel bookings, employee payroll, and beneﬁts

processing are typically outsourced. Even ﬁnancial ac-

counting and tax preparation tasks get outsourced in

typical small ﬁrms. Call centers and help desks are

outsourced routinely these days. Outsourcing business

processes to offshore service providers is a growing

trend [71.16]. BPO can result in signiﬁcant operational

beneﬁts as documented through many case studies in

the BPO Journal [71.17]. Processes are typically inter-

woven with computer and manual activities. As an il-

lustration, a help desk process is illustrated in Fig.71.2.

It can be seen that the human processes are interwo-

ven with machine processes. The human processes call

for stand-alone tasks by the service provider as well as

interactive tasks with the customer. Periodic reference

to adatabase is needed. Often adecision hasto be made,

with or without the help of the machine.

Telephone call received from

a customer

Seek customer feedback,

complete system log

Communicate to the

customer

Transfer the con-

versation to an

expert for resolution

Refer to system to seek

resolution if issue is of

known causes

Interact with the customer to

authenticate customer bonafide

Interact with the customer to

authenticate the person

Interact with the computer to

validate service eligibility

Interact with the customer to

learn about the issue

Determine prima facie if the

issue is within the service

agreement (use system)

Regret

service

Negative

response

Negative

response

Fig. 71.2 Help desk process

Processes where there is no interaction with the cus-

tomer are called back ofﬁce and the others are known as

front ofﬁce. The outsourcing opportunity arises due to

many factors.

The volume of calls received by a particular ﬁrm

or product division may be insufﬁcient to invest in

a system or to maintain a help desk internally without

substantial idle time for the resources. A third-party ser-

vice provider can be more efﬁcient when its resources

are shared across multiple companies. Substantial in-

vestment in domain expertise needed is yet another

reason for outsourcing; for example, payroll process-

ing for a ﬁrm operating in many states calls for current

knowledge of state-level taxes and other labor rules and

legislationand incorporating them in the systemwith no

time delay.

It is common practice to hand over – lock, stock and

barrel – the existing system (hardware and software),

people, and processes to the outsourced vendor to gain

business efﬁciency. Firms ensure the efﬁciency gains

and continuing improvements through service-level

agreements (SLAs). SLAs have become sophisticated

in design but simple in implementation over the years.

They specify the level of service satisfaction to be

reached for different processes. Further strict control

over accountability is established.

Vendors ﬁnd numerous opportunities to automate

and gain efﬁciency along the way. In the help desk ex-

ample cited above, the ﬁrst opportunity is to log all

incoming calls automatically and sort them by probable

cause; then to build the frequently encountered issues

and their resolution database and add a voice-based sys-

tem to guide the customer through resolution of simple

Part G 71.3

1272 Part G Infrastructure and Service Automation

issues. Expertise gained over the help desk system can

be converted into frequently asked questions (FAQs)

and incorporated into the online help features of the

system, thus minimizing the need to call. The vendor

ﬁrm can even share knowledge gained from one cus-

tomer ﬁrm environment to others by using a common

high-level issues-resolution team,if agreeableto allcus-

tomers. Further periodic analysis of issue logs would

reveal what issues are encountered most often and why.

Then it becomes a valuable data source to modify or

enhance the associated manual processes or the embed-

ded software. It is not uncommon for ﬁrms to outsource

business processes to standardize or streamline them

across many segments when they are reorganizing or

consolidating.

Many tools are available in the market to support

BPO ﬁrms and their clients.

71.3.4 Analytics

Decision support systems (DSS) have played a promi-

nent role in widespread adoption of computers from the

early decades. As opposed to the challenges of storing

and processing large volumes of data in transaction pro-

Online analytical

processing:

pattern recognition:

search engines

Large-scale

optimization:

animation

Nonlinear

programming:

stat. analysis:

simulation

Linear

programming;

regression

Fig. 71.3 Growth in analytical applications

cessing systems (TPSs), DSS have been computation

intensive. See Chap.87 for additional content on deci-

sion supportsystems. While operational researchers and

other management science experts have been successful

in mathematically modeling large business and engi-

neering systems even in the 1950s, the implementation

of thesesystems had to await advancements inhardware

and software. Dantzig’s simplex algorithm, invented in

1947, to solve linear programming problems and the in-

ventionof computerin the late 1940s can be taken asthe

starting points. It took another two decades before prod-

uct mix optimization in reﬁneries, route optimization

and crew and vehicle scheduling in mass transporta-

tion systems, production scheduling, and many other

large-scale problems could be solved in reasonable time

periods (within an hour or much less) to be of practical

value.

Inventory management, project management sys-

tems using program evaluation and review tech-

nique/critical path method (PERT/CPM) methodology,

network ﬂows optimization, and time series and re-

gression analysis were the solutions implemented using

powerful mainframe computers until the early 1970s.

This period also saw the birth of discrete event and con-

Part G 71.3

Automating Information and Technology Services 71.3 Automation Path in Each Business Segment 1273

tinuous simulation techniques. They found widespread

acceptance in the industrial world.

The ever-increasing processing power of comput-

ers coupled with their ability to store vast amounts of

data served to fuel the growth of analytical applications.

Very large-scale optimization problems involving hun-

dreds of thousands of decision variables and constraints

could be taken up initially with supercomputers such

as those built by Cray. Similarly Silicon Graphics in-

troduced the power of the computer into the world of

animation, which meant that high-end engineering sys-

tems can now be simulated and visually represented on

computer screens for better understanding and use.

Another remarkable shift has been the use of per-

sonal computers in analytics. Numerous tool vendors

have exploited the exponential growth in computers’

processing power to develop PC-based solutions, CPlex

from Ilog, Frontline Solver, MSProject, SPSS, and SAS

for statistical analysis to name a few. Similarly one can

mention Arena and Automod in the simulation world.

One of the signiﬁcant developments during the

1980s was the emergence of yield and revenue man-

agement concepts in the airline industry. Analyzing

millions of transaction data from past ﬂights, the team

of operations research (O.R.) experts at American Air-

lines was able to come up with a ﬁne-tuned strategy

for pricing airline seats and dynamically altering them

until ﬂight departure. Success of these techniques en-

couraged the entertainment and hospitality industry,

consisting of hotels, vacation travel, and recreational

sports, to embrace these concepts and improve their

proﬁtability.

Corporate data usually resides in multiple databases

in various formats. Recognizing the need to bring them

under a common framework in a uniﬁed database, ana-

lytics vendors have ﬂoated data warehousing tools. Data

mining solutionsthat can extract statistically valid infer-

ences from these data warehouses have also emerged.

Analytical processing has gone online during the

past decade. Online analytical processing (OLAP) tools

from many vendors enable ﬁrms to perform analyt-

ics in real time and incorporate the decision variables

into real-time control systems. Pattern recognition and

search engines have been integral to many web-based

applications. Google’s success is to be credited to the

optimized page-ranking algorithm tied to its search en-

gine. Figure 71.3 traces the growth of analytics over the

past decades.

Analytics is poised to play a signiﬁcant role in

corporate competitive advantage, as outlined by Dav-

enport [71.18]. It can be a powerful tool in providing

transparency, rooting out gray markets, reducing cor-

ruption, and being fair to all stakeholders where

judgment is used to select some for negative treatment.

This is covered by Balasubramanian [71.19].

71.3.5 Printing and Display Solutions

During their inception, computers relied on character

display terminals and line printers. Computer output

was either viewed on the terminal or printed. Mayadas

et al. [71.20] have tracked the growth of this industry

until the mid 1980s. During the batch processing days

line printers were installed in computer centers. The op-

erator had to insert continuous stationery, print multiple

copies, and sort the output for dispatch and distribu-

tion. The principal issue to be tackled was the mismatch

between the processing speed of the computer and the

speed of the printer. Research efforts were focused on

enhancing the printing speed step by step.

Dot matrix printers (DMPs) were introduced in the

1980s, with the signiﬁcant innovation that a printed

page could be either portrait or landscape style. It even

became possible to print pictures, albeit at low levels

of accuracy. DMPs were priced low and they cap-

tured the PC-based systems market aggressively. Inkjet

and laserjet printers were introduced in the 1990s and

have revolutionized the printing space. Systems are now

free from the need for continuous stationery too. Color

printing became common very quickly and the desktop

printing (DTP) segment has evolved as a distinct area.

With theemergence of DTP, preparation of high-quality

slides for corporate presentation, university research re-

ports containing mathematical symbols, and artwork on

tee shirts all became possible.

Very soon printers became intelligent devices with

the help of software. They could hold a line of printing

jobs, schedule them based on priority, change fonts and

features from one job to another, report paper or ink car-

tridge issues,take care of reverse printing, and eliminate

human effort and errors. Postscript language and digital

fonts were developed by Adobe Systems in 1982, thus

facilitating drawings and image manipulation. Printers

capable of engineering drawings and blueprints were

not to be left behind. So are other innovations such as

photo-printing machines, which are now ubiquitous. In

the Internet age, network printers can act as copying or

faxing machines as well.

The technology-lead changes in the computer

screen market are equally impressive [71.20].Graphical

and iconic display is taken for granted now. However, it

is a far cry from the character display terminals. Mov-

Part G 71.3

1274 Part G Infrastructure and Service Automation

ing from cathode-ray tube (CRT) to light-emitting diode

(LED) to plasma screens, computer system users have

been empowered to display sophisticated images of

simulated vehicle crashes, human organs, ocean ﬂoors

at depth etc. along with traditional business graphs,

tables, and legal documents. The ﬁeld of data visual-

ization is merging graphics and animation to enable

dynamic, real-time perception and pattern recogni-

tion [71.21].

The need for an intermediary to analyze output,

interpret results or initiate control action in process con-

trol systems has been minimized by the integration of

computer-based systems. The console is a device for

oversight, otherwise all controlling tasks can be done

by thesystem. Automation from end to end is thus being

enabled.

One of the visible trends for the future is continu-

ing eliminationof printed paper. Acceptanceof digitally

ﬁled tax returns, electronic voting, and electronic mail

as legally valid communication are good examples. Ad-

vancesin digital certiﬁcation technologyplay a catalytic

role here [71.22].

71.4 Information Technology Services

71.4.1 Computer-Aided

Software Engineering

There are four distinct stake holders in information sys-

tems, namely, system users (company executives), their

subject of usage (such as factory operations), computer

system operators, and system developers. Their expec-

tations and requirements are not concurrent and are

usually diverse. The responsibility to design the system

to meet every stakeholder’s needs, however, lies with

the developers [71.23].

The key stages of software engineering are require-

ments analysis, functional speciﬁcation preparation,

architecture and system design, software development,

testing, implementation, maintenance, and documenta-

tion. Theseare manpower intensive and were performed

without the help of any tool to begin with. This scenario

was to change soon.

Software engineering was the focus of the Defense

Advanced Research Projects Agency (DARPA)andthe

National Institute of Standards and Technology (NIST)

even in the formative years of computers [71.24]. How-

ever it soon became a mission critical issue for system

developers such as IBM, EDS, Accenture, and HP.

Confronted with the twin challenges of improving the

quality of software and enhancing the productivity of

software developers, many service providers turned to

streamlining the processes, to adopting quality assur-

ance methodologies, and to the tools of their trade,

seeking solutions. They recognized that automating

many life cycle activities of software development can

bring immense beneﬁts internally. This was, however,

easier stated than accomplished.

The common business-oriented language (COBOL)

and Fortran were the dominant languages for program-

ming for business and scientiﬁc applications until the

1970s. By developing newer languages that were fea-

ture rich, that could handle multiple data types, and that

were easier to learn and to deploy the industry sought to

address itsconcerns. It was a daunting challenge to keep

up with the evolution in database management [71.15]

concepts too. A cherished dream of software develop-

ers to reuse code became a possibility, initially with

subroutines and libraries and later with the invention

of object-oriented methodology [71.25] and languages

such as Smalltalk, Lisp, and C++.

Run-time efﬁciency is also a major concern with

many systems. How does one ensure that only rel-

evant sections of the code get executed without the

need to carry along redundant code? This issue sur-

faces particularly when a developed system is striving

to meet the functionality requirements of many cus-

tomers with partially nonoverlapping requirements.

Dynamic languages such as Perl, JavaScript, PHP,

and Smalltalk provide this run-time ﬂexibility. Hence,

choosing the appropriate language becomes a judicious

decision [71.26].

Efforts analysis of many major projects revealed

that nearly 70% of resources are consumed by the pre-

and post-coding stages of the software development life

cycle. Besides, annual efforts expended to maintain the

software were found to account for around 15% of the

development stage. It dawned on the service providers

that it is imperative to focus on both the pre- and post-

coding stage activities also.

Design tools such as DesignAid, Excelerator,

IEW, ADW, IEF, and Accenture’s Foundation toolset

(METHOD/1, DESIGN/1, etc.) became popular as they

helped to streamline and standardize the design process,

to serve as a means of communication with the end

users, and to document the system. While many tools

were on the PC platform, Unix and mainframe-based

Part G 71.4