_ «..J. a.. en en. Information Systems Planning Guide U') GE ::J ' . 0

Transcription

1 z z «..J a.. en :E w - en en U') w z en ::J ' _.- nformation Systems Planning Guide GE

2 CJ z z Z <C..J C '_ -- - nformation Systems Planning Guide en en w z en ::l m Business Systems Planning (BSP) is a strutured approah to help an organization establish an information systems plan that an satisfy its near- and long-terminformation needs. This guide offers assistane to BSP study teams by explaining the approah and presenting guidelines for onduting a BSP study. t an also be used for general referene on the subjet of planning for information systems. The guide overs BSP priniples, onepts, and study overview, then devotes a hapter to eah of the major ativities of a BSP study, and inludes a fmal hapter to aquaint the reader with follow-on projets.

3 Seond Edition (Otober 1978) A form for readers' omments has been provided at the bak of this publiation. f the form has been removed, address omments to BM Corporation, Tehnial Publiations, Dept. 824, 1133 Westhester Avenue, White Plains, New York BM may use or distribute any of the information you supply in any way it believes appropriate without inurring any obligation whatever. Copyright nternational Business Mahines Corporation 1975, 1978

4 Prefae Business Systems Planning (BSP) is a strutured approah to assist a business in establishing an information systems plan to satisfy its near- and long-term information needs. The primary purpose of this guide is to assist BSP study team members by explaining the approah and presenting speifi guidelines for onduting a BSP study. The guide may also be used for general referene on the subjet of planning for information systems. The reader does not need a knowledge of omputers to understand the onepts and to apply the methodology. Experiene has shown that BSP an be applied to all the publi setor and all industries in the private setor, beause the requirements for developing information systems are similar regardless of the business served or the produts and servies provided. F or the sake of simpliity, throughout this manual the term business is used for the organization entity being studied, regardless of its size or purpose, and whether private or publi. Before initiating a BSP study, the personnel involved should onsider taking the appropriate BSP-related ourses. nformation on the eduational offerings is available through loal BM branh offies. Beause BM annot ontrol the manner in whih this methodology is applied, it takes no responsibility for the results. Hundreds of studies have been ompleted suessfully using this approah; this guide represents as muh of that experiene as it is feasible to inlude. This guide is organized so that readers already familiar with BSP an go diretly to the material they require, while others may progress serially through priniples, onepts, methodology overview, methodology details, and finally the alternati and the referene material in the appendies. The ittrodution is for those persons who may have heard of BSP but do not know its purpose or origin. Chapter presents the onepts that form the basis for the methodology, so that the reader an understand the basi priniples without beoming embroiled in the detail of how BSP is done. Even persons with some orientation or training in the subjet may wish to read the introdution and Chapter to strengthen their understanding. Chapter 2 provides an overview of all the ativities in a BSP study so that the reader will be prepared for the details in the following hapters. Chapter 3 deals with getting the proper exeutive ommitment before starting a BSP study. Chapter 4 disusses the ativities that take plae before the BSP team starts its onentrated study. t also outlines the onsiderations that must be made in the management of the study. Chapters 5-15 over the details of the methodology, ulminating in the writing of the final report and the making of reommendations to the business exeutives. Sine eah step of the methodology an prohably be done a number of ways, a single, ohesive methodology is outlined and alternatives, where appropriate, are inluded at the end of eah hapter or in the appendies. For ontinuity from step to step, all examples are based on the same business, a mediumsize manufaturing firm, exept those in the appendies, whih are based on various industries and the publi setor. Chapter 16 gives an introdution to the follow-on projets so that an ation plan for them may be developed during the BSP study.

5

6 Contents ntrodution The Changing Environment From Operation to Management Control A Justified Approah Origin of BM's Business Systems Planning Funtion 2 Objetives and Potential Benefits of BSP 2 Objetives 2 Potential Benefits 3 Chapter t. BSP Conepts 4 Chapter 2. Overview of the BSP Study Approah 10 Perspetive '10 Major Ativities 11 Gaining Commitment 12 Preparing for the Study 12 Conduting the Kikoff Meeting 12 Defining Business Proesses 12 Defining Data Classes 12 Analyzing Business/Systems Relationships 13 Determining the Exeutive Perspetive 13 Assessing Business Problems 13 Defining nformation Arhiteture 13 Determining Arhitetural Priorities 13 Reviewing nformation Systems Management (SM) 13 Developing Reommendations and Ation Plan 13 Reporting Results 13 Chapter 3. Gaining the Commitment 15 Establishing the Study Sope 15 Setting Study Objetives 16 Developing Business Reasons for the Study 16 Resouring the Study Team 16 Charateristis 16 Organization and Responsibilities 16 Considerations for Full-Time or Part-Time Study 17 Manpower Requirements 17 Seletion of Team Leader 17 Sponsor's Letter 17 Chapter 4. Preparing for the Study 18 Briefing the Study Team 18 Gathering Data 18 Eduating the Team 19 Developing the Study Work Plan 19 Establishing a BSP Study Control File 21 Defining the Struture of the Final Study Report and Anillary Output 21 Creating Exeutive nterview List and Shedule 21 Briefing nterviewees 21 Obtaining and Equipping Control Room 22 Arranging for Administrative Support 22 Reviewing with Exeutive Sponsor 22 Chapter 5. Conduting the Kikoff Meeting 23 Sponsor's Opening 23 Review of Study Work Plan 23 Business Review 23 nformation Systems Review 23 Chapter 6. Defining Business Proesses 24 Prerequisites to Defining Proesses 24 The Produt and Resoure Life Cyle 24 Basi Steps in Defining Proesses 24 Planning and Control Proesses 25 Define the Proesses of Strategi Planning and Management Control 25 Produt/Servie Proesses 26 dentify the Produt/Servie of the Organization 26 dentify the Proesses in Eah Stage of the 'Life Cyle of the Produt/Servie 26 Write a Desription of Eah Proess 26 Supporting Resoure Proesses 28 Desribe the Supporting Resoures 28 dentify the Proesses in Eah Stage of the Life Cyle of Eah of the Supporting Resoures 28 Consolidation and Analysis 29 Group the Proesses 29 Chart the Proess Groups and Complete Desriptions 29 Relate the Business Proesses to the Organization 29 dentify Proesses Key to Business Suess 30 Outputs and Their Uses 30 Expansions and Variations in Approah 30 Diversity in Business, Produt, or Servie 30 Outside Assistane 30 Alternate Methods in dentifying Business Proesses 32 Generi Proess Model 32 Chapter 7. Defining Data Classes 33 Business Entity Approah 33 Business Proess Approah 35 Cross-Referening and Regrouping 35 Data Class to Proess Analysis 36 Chapter 8. Analyzing Business/Systems Relationships 37 Understanding l/s Support of Proesses 37 dentifying Usage of Current Data 41 Summary 41 Chapter 9. Determining the Exeutive Perspetive 42 Confirming the List of Exeutives to be nterviewed 42 Reviewing the nterview Shedule 42 Preparing Questions for nterviews 42 Determining Team Member nterview Roles and nterviewee Charateristis 43 Conduting nterviews 44 Preparing nterview Summaries 44 Chapter to. Assessing Business Problems and Benefits 45 Step : Summarize nterview Data 45 Step 2: Classify nterview Data 45 Step 3: Relate Data to Proesses 47 Summary 47 Chapter t 1. Defining nformation Arhiteture 49 Developing the Arhiteture Diagram 49 Data Flow 53 Subsystem dentifiation 57 Analyzing for Prerequisites 57 Arhiteture Use 57

7 Chapter 12. Determining Arhiteture Priorities 58 Determine Seletion Criteria 58 Apply Criteria and List Systems 58 List Prerequisites 58 Doument Reommended System 58 Expansions and Variations 59 Chapter 13. Reviewing nformation Systems Management 60 Purpose of SM in the BSP Study 60 SM as Part of the BSP Study 60 SM nputs 60 Flow of Ativities in the SM Module 60 Compiling the /S Objetives 60 Ensuring SM Supports Arhitetural Priorities 62 Problem Analysis 62 Rearrange /S Support Problems by Type 62 Extrat SM Problems from /S Review 62 Consolidating and Prioritizing SM Changes 62 Developing Reommendations and Ation Plan 63 SM Outputs of a BSP Study 63 Chapter 14. Developing Reommendations and Ation Plan 64 Chapter 15. Reporting Results 65 Report Outline 65 Report Preparation 65 Presentation Medium 65 Exeutive Presentation 65 Chapter 16. Overview of Follow-on Ativities 66 Perspetive on Follow-on Ativities 66 Relation of BSP Study to Follow-on Ativities 66 Preparation for Follow-on Ativities 66 Orientation of Teams 66 nformation Systems Management 67 SM in Perspetive 67 nformation Systems as a Business 67 nformation Arhiteture 68 Arhiteture Refinement 68 Current Systems Examination 69 Data Base 69 Distributed nformation Systems 69. Developing the First System 72 Tehnial Fators 72 Effetive Communiation 72 Priorities and Conflits 72 Appendix A. Sample Exeutive Announement Letter 74 Appendix B. Sample nterview Confirmation Letter 75 Appendix C. Examples of Proess Groups and Proesses by ndustry 76 Appendix D. Examples of Planning and Control Proesses 81 Appendix E. Sample Desriptions of Proesses 83 Appendix F. Examples of Business/Systems Relationship Matries 86 Appendix G. Sample List of Wall Charts for Control Room 89 Appendix H. Risk - Potential Benefit Analysis 90 Appendix. SM Proesses 92 Appendix J. Potential Topis for BSP Study Report 94 Glossary 96 Figures. Translation of business strategy to /S strategy 4 2. Charateristis of planning and ontrol levels 5 3. Top-down analysis with bottom-up implementation 7 4. General /S planning approah 8 5. nformation arhiteture example 9 6. Relationship of BSP to /S projets Flow of the BSP study 8. BSP organizational business unit options Sample study team organization BSP study ontrol points 19. Work plan for the BSP study Definition of business proesses Example of the flow ofa produt/servie through a business 14. Sample list of proesses by proess group Organization/proess matrix Business proess network nformation life-yle Data lass/business entity matrix nput-proess-output examples Data lass by proess, showing data reation and usage Organization/proess matrix, with urrent systems support shown System/organization matrix System/proess matrix System/data file matrix Data redution steps nterview analysis and data redution worksheet Problem/proess matrix, showing the number of times the problems were stated Data lass by proess, showing data reation and usage Data lasses arranged by reating proess Proess/data lass groupings Data flow determination Data flow nformation arhiteture Graphi rearrangement of information arhiteture Prerequisite subsystem analysis Sample subsystem ranking Flow ofsm in a BSP study Data base system Distributed information requirements analysis 71

8 ntrodution The Changing Environment The rapidly hanging environment and the onstant need for businesses to adjust quikly to it make it neessary for exeutive management to have up-to-date information avaiiable at all times, so that through meaningful analyses and resoure alloation tradeoffs they an manage their businesses more effetively. With organization-wide availability of information, strategies an be improved, deisions made more soundly, and operations performed more effiiently. From Operation to Management Control Data proessing is in transition. Until omparatively reently, most businesses onsidered it a servie funtion and used it to support single-operation units or loations. n the last two deades appliations have been developed independently, with very little regard for the support they ould give other funtions and for the information they ould supply for management ontrol. Funtional autonomy has been the rule. This has resulted in frationalized and redundant data files and in the inaessibility of data from the many operational appliations installed in the various funtions. Today, however, businesses are reognizing data more and more as a resoure that is as important as personnel, ash, failities, or materials. They see the need to onsolidate the key data files and make information available not just to individual funtions or departments but throughout the business, in order, for management to gain a business-wide view and be able to make multifuntional deisions. Many ompanies have reognized the need for ompany-wide information systems but have been unable to develop them for one or more of the following reasons: Failing to obtain exeutive ommitment and involvement Establishing objetives and strategies that were not in line with their overall business objetives Attempting to implement information systems without first understanding the business from general management's viewpoint Setting out to implement totally new ompany-wide information systems rather than a omprehensive plan evolving from urrent systems failing to put in plae-those information systems management funtions required to adequately manage the information systems resoures A Justified Approah n an artile entitled "Blueprint for MS," * Dr. William Zani states, "Traditionally, management information systems have not really been designed at all. They have been spun off as by-produts while improving existing systems within a ompany. No tool has proved so disappointing in use. trae this disappointment to the fat that most management information systems have been developed in the "bottom-up" fashion - an effetive system, under normal onditions, an only be born of a arefully planned, rational design that looks down from the top, the natural vantage point of the managers who will use it." n most instanes the urrent operational systems have been justified and have been performing effetively for their speifi intent, even though their maintenane and interfaes may have beome unmanageable. An information system plan should allow a modular approah to implementation, providing onfidene that eah module will fit and funtion properly in an integrated network and will interfae properly with the present operating systems until they too an be inluded in that network. The plan should also allow for better deisions onerning the effiient and effetive ommitment of information systems development resoures. With suh a plan, the required information an be more readily obtained. Business Systems Planning (BSP) is geared to help provide suh a plan through: A top-down approah to (1) getting people ommitted and involved (starting with top management and working down through the organization) and (2) studying the business (working from the overall to the detail level) A bottom-up approah to implementation Use of a strutured methodology proven in hundreds of studies The translation of business objetives into information requirements The effetiveness of the BSP methodology an be attributed to two omponents: The fundamental priniples and onepts - the unvarying ideas and logi that form the basis for BSP, inluding the standards upon whih the proedures are based *Harvard Business Review, November/Deember,

9 The sequened ativities, tehniques, disiplines, time,outputs, planning, team omposition, et., established to fill a partiular organization's need and situation (although onsistent with BSP's basi priniples and onepts, the pr6edures are flexible and vary with the partiular environment) Origin of BM's Business Systems Planning Funtion Learning from its own mistakes and those of other ompanies that attempted to implement large information systems in the 1960s, BM realized that a disiplined approah was required, using proven priniples and methodologies. n 1966 a business-wide nformation Systems Control and Planning Department was established at BM'S Data Proessing Group headquarters. The Data Proessing Group was a total business unit omprising the engineering, manufaturing, marketing, and servie divisions responsible for all of BM'S domesti data proessing business. Until the Control and Planning Department was established, BM had little overall diretion in the internal use of omputers. n fat, little oordination took plae between divisions; most data proessing ativities were onfined to loations and units within divisions. Consequently, eah manufaturing plant and marketing region developed and operated its own system. Although the individual systems arried out similar funtions, they differed in design and performane; they ould not be used interhangeably and ould not ommuniate with eah other. The result was a redundany of data and exessive use of th data proessing resoures required to develop and maintain suh systems. Even with this large expenditure of resoures by eah division, the systems were mainly satisfying the loal department needs of the business, 2 rather than doing an overall data proessing job. When steps were taken to improve the data proessing within a division (for example, development of a onsoli,dated order entry system within the marketing division), little serious attention was given to an effetive interfae of that system with inputs from engineering and outputs to manufaturing. The business was not getting the return on investment from data proessing that it ould have beause the information needs of the business, and partiularly those of the generalll)anager responsible for the business, were not being aommodated. The first effort of the Control and Planning Department was to inventory and profile the systems existingwithin the business and the plans for the future. At the same time, reognizing that the data proessing effort must be direted toward satisfying business needs and not solely toward individual funtions and departments, the Control -and Planning Department established a set of information system strategies overing five major areas: 1. Fixed data responsibility 2. Single soure and parallel distribution of data 3. Central ontrol and planning of information systems 4. Organizational independene of data 5. Resoure sharing of data, equipment, and ommuniations With the knowledge of what was being done with data proessing, and the diretion established through the set of strategies, the department defined a network. of information systems and assigned responsibilities for the development of the systems. These systems addressed the operational, funtional, and general management needs for information. As the definition and design efforts for the businesswide network of information systems got under way in the late 1960s, many of BM's ustomers showed interest in learning how they might better manage their information system resoures. n an effort to assist these interested ustomers, BM established the Business Systems Planning (BSP) program in To ondut the program, the nuleus of the ontrol and planning department that developed the internal information systems plan was transferred to the Data Proessing Division headquarters. This group proeeded to doument proven methodologies and institute a training program to eduate regional and branh offie personnel and ustomers in the approah. Exeutive briefings and seminars were established to show ustomer exeutives the potential benefits of the approah and the reasons why their involvement was vital to suessful implementation of information systems. The methods used in developing this information systems plan and the lessons learned have sine been used by many BM ustomers. Appliation of BSP methodology has helped them to formulate their information system plans and ontrol mehanisms and to improve their use of information and data proessing resoures. Studies using the BSP methodology and guidelines have been onduted suessfully by profit and nonprofit organizations, of varying size, in many industries. Otives and Potential Benefits of B - With a reasonable amount of planning and ontrol, the user of this BSP approah should be able to attain its objetives and realize its potential benefits. Objetives The first and most important objetive of BSP is to provide an information systems plan that supports the business's short- and long-term information needs and

10 is integral with the business plan. There are other objetives that help to justify and larify the approah: 1. Provide a formal, objetive method for management to establish information systems priorities without regard to provinial interests. 2. Provide for the development of systems that have a long life, proteting the systems investment, beause these systems are based upon the business proesses that are generally unaffeted by organizational. hanges. 3. Provide that the data proessing resoures are managed for the most effiient and effetive support of the business goals. 4. nrease exeutive onfidene that high-return, major information systems will be produed. 5. mprove relationships between the information systems department and users by providing for systems that are responsive to user requirements and priorities. 6. dentify data as a orporate resoure that should be planned, managed, and ontrolled in order to be used effetively by everyone. The illustration below shows the overall flow of BSP and relates the BSP study to the ativities that follow. This relationship is expanded upon in Chapter 16, with an explanation of the foll()w-on ativities. Step Evaluation of BSP Exeutive ommitment Preparation for BSP study BSP study Output Analysis Sope of study Objetives Team leader Study team Eduation Work plan Business fats Business requirements nformation arhiteture Arhitetural priorities lis management Ation plan Potential Benefits Appliation of the approah and methodology ontained in this planning guide offers many potential benefits to three management groups: To exeutive management: An evaluation of the effetiveness of urrent information systems A defined, logial approah to aid in solving management ontrol problems from a business perspetive An assessment of future information system needs based on business-related impats and priorities A planned approah that will allow an early return on the ompany's information systems investment nformation systems that are relatively independent of organization struture Confidene that information system diretion and adequate management attention exist to implement the proposed systems To funtional and operational management: A defined, logial approah to aid in solving management ontrol and operational ontrol problems Consistent data to be used and shared by all users Top management involvement to establish organizational objetives and diretion, as well as agreedupon system priorities Systems that are management and user oriented rather than data proessing oriented To information systems management: Top management ommuniation and awareness. Agreed-upon system priorities A better long-range planning base for data proessing resoures and funding Personnel better trained and more experiened in planning data proessing to respond to business needs The plan that results from a BSP study should not be onsidered unhangeable; it simply represents the best thinking at a ertain point in time. The real value of the BSP approah is that it offers the opportunity to (1) reate an environment and an initial plan of ation that an enable a business to reat to future hanges in priorities and 'diretion without radial disruptions in systems design, and (2) define an information system funtion to ontinue the planning proess. Follow-on projets nformation arhiteture First system(s) ls management Data management ls planning 3

11 Chapter 1. BSP Conepts Business Systems Planning is most often thought of as a strutured approah or methodology. This methodology, however, is based upon some fundamental onepts, a good understanding of whih an give the BSP study team members:. A better appreiation of the ';;why's" of the methodology 2. mproved onfidene in applying variations to meet speifi situations 3. A better bakground with whih to ommuniate the objetives and eventual reommendations to senior management The premise for onduting a BSP study is that there exists witb.in the organization a need for signifiantly improved ]omputer-based information systems (lis) and a need for an overall strategy to attain them. BSP is onerned with how these information systems should be strutured, integrated, and implemented over the long term. The basi onepts of BSP an be related to the long-term objetives for lis in an organization. An ls Must Support the Goals and Objetives of the Business This most basi onept underlies the "top down" philosophy of the methodology as well as several of the speifi steps, suh as exeutive interviews and system priorities. Sine information systems an be an integral part of a business and be ritial to its overall effetiveness, and beause they will ontinue to represent major investments of time and money, it is essential that they support the organization's true business needs and diretly influene its objetives. BSP, then, an be thought of as a vehile or proess to translate business strategy into lis strategy (see Figure ). Business Strategy Mission Goals and objetives Strategies BSP /S Strategi... Planning Proess... /S Strategy /S Objetives /S Poliies nformation Arhiteture Figure 1. Translation of business strategy to lis strategy Obviously, it is important that an organization be willing and able to express its long-term goals and objetives. For some organizations, this an be done prinipally through the business plan. For others, where a business plan is not available or urrent, it an 4 be done as a part of the BSP methodology. n either event, a reognition of this basi need by senior management is ritial, for only with that reognition will their ommitment and involvement be great enough to guarantee a meaningful BSP study. An ls Strategy Should Address the Needs of All ",. ' Levels of Management Within the Business This requirement has several impliations relative to lis struture. First, it is important to reognize the varying harateristis of information as needed by different ativities and management levels. Typially, lower levels need onsiderable detail, volume, and frequeny, higher levels need summaries, "exeption" reporting, and inquiries, and still higher levels need ross-funtional summaries, speial requests, "what if' analyses, "external" requirements, et. t would be impratial to onstrut a single system to aommodate all ativities or management levels, and it would be erroneous to assoiate anyone type of information requirement solely with one management level. Clearly, there is need to establish some reasonable framework upon whih the lis an be defined. First, the emphasis in lis should be in support of management deision making. This is in ontrast to more traditional bookkeeping or reordkeeping funtions. Business deisions are made for various purposes, but most an be assoiated with either planning or ontrol. Planning, of ourse, is the establishment of various missions, objetives, and poliies, and it ours at all levels. Good information is vital to the establishment of good plans. Control deisions, on the other hand, are made in order to guide an ativity toward some impliit or defined (by the plan) objetive. The lis an provide the measurements of the urrent or atual ondition to the deision maker. We thus omplete a planning, measurement and ontrol yle with lis potentially an integral part. Sine planning and ontrol are the keys to deision making, a framework for lis based upon these ativities an be utilized. t has been proposed, * and well aepted today, that three distint but onurrent planning and ontrol levels exist in any organization: Strategi planning, the proess of deiding on objetives of the organization, on the resoures used to attain these objetives, and on the poliies that are to govern the aquisition, use, and disposition of resoures *Anthony, R.N. Planning and Control Systems: A Framework for Analysis, Harvard Business Shool, Division of Researh, 1965.

12 Management ontrol, the proess by whih managers assure that resoures are obtained and used effiiently in the aomplishment of the organization's objetives Operational ontrol, the proess of assuring that speifi tasks are arried out effetively and effiiently Further harateristis of these areas are outlined in Figure 2. An advantage of this framework is that it does not restrit planning and ontrol ativity to any partiular industry, funtion, or management level. A onlusion at this point is that an ljs ould onveniently address itself to anyone of the above three planning and ontrol levels. Resoure management is also key to this philosophy and represents a major vehile for US definition. The speifi resoures to be managed vary in nature and relative importane from one organization to the next. Examples of traditional resoures to be managed are people, failities, materials and money. Their requirements are based upon the needs to support the prime mission area of the organization, e.g., its produts or servies. Beause an organization's produt or servie area has all the attributes of a resoure, i.e., a life yle of ativities and deision points, and yet drives the other resoures, it is referred to as the "key resoure." Eah resoure, inluding the key resoure, is managed through planning and ontrol deisions of the three levels previously disussed. Resoure management has the desired harateristi of utting aross organizational boundaries - vertially aross management levels and horizontally aross funtional lines. Thus a framework based on resoures as well as planning and ontrol levels an be established, and an US arhiteture an be applied within this framework. An lis Should Provide Consisteny of nformation Throughout the Organization The keyword in this objetive is onsisteny. The impliation is that information derived from more traditional data proessing appliations is not neessarily onsistent, partiularly when applied to new business problems (deision areas) of broader sope. The problems in data onsisteny normally arise as a result of a historial evolution of omputer usage that traditionally ours. solated and independent appliation areas are seleted and mehanized, typially to redue operational osts. The'data files are defined as neessary to support the speifi needs of eah appliation without regard to one another or to future appliations. The data itself is onverted from manual files loated and maintained by the using organization. As omputer appliations are added, new data files are usually required sine the data requirements for different appliations are rarely the same. These are usually reated from spinoffs of existing mehanized files plus any additional data that may be required from the using area. Summary-type reports for higher management levels are the result of sorting and merging various existing data files together to reate new ones. Rarely is any existing data file of the form or ontent required to provide newly requested information of any magnitude. Thus new data files are born. Data redundanies and file update requirements multiply. The ontinual ry from management that " know the data that need is somewhere in data proessing, but an't get at it" may be basially true from their point of view. The data may be there, but not neessarily defined as needed, of adequate summary level or sequene, or of proper time period or urreny. Planning and Control Level Deision Strategi Management Operational Charateristi Planning Control Control Management General management General management Funtional management. involvement Funtional management Funtional management Operational management Operational management Time Long range Vear-to-year Day-to-day horizon (1-10+ years) Monthly Weekly Degree of Unstrutu red and More strutured, Highly strutured, struture irregular; eah problem yli, largely repetitious different repeating Data Summaries, estimates, dif- Summaries, definable, need Detail, operational, definrequirements fiult to pre-define, muh for unantiipated forms, able, internally generated external to business largely internal Resoure Establishment of poliies Alloation of the resoure Effiient use of the management pertaining to the resoure resoure Figure 2. Charateristis of planning and ontrol levels 5

13 Data, then, exists in most organizations in varying form, definition, and time. The form may be unaptured raw data, mehanized data files, detailed DP reports, summarized DP reports, business douments, or knowledge in someone's head. The definition of any given data an have as many variations, and thus inonsistenies, as there are users of that data. For example, "salary" to the payroll department may mean an employee's atual monthly pay, to the projet manager an annual figure plus burden to be harged to a ustomer, to the department manager a budget line item representing total expense for all reporting employees. n addition, a time inonsisteny is very likely to exist between what may otherwise be omparable data. Data may be aptured by suh varying methods as the mail, telephone, data terminals, or satellites. t may be "bathed" over varying lengths of time by the user or by OP before proessing, or it may be entered "online" as eah transatiorl ours, diretly from its soure. Data may be proessed daily, weekly, or monthly on a predetermined shedule, or it may be inapable of being proessed until a series of prior omputer runs are ompleted (e.g., the month-end losing). The output itself may be mailed or it ould be immediately available as the result of an online inquiry. Finally, the report may be up to the seond, as when reading from a terminal, or it may have lain in a desk for three weeks after last month's proessing. The ombinations of time deviations between data apture, data proessing ativity, and atual data usage are many. With all these potential data inonsistenies, it is no wonder that reports frequently don't "math" between using departments and managers. This beomes a problem most often during interdepartmental deisionmaking or at higher reporting levels where onsolidation of multifuntion ativities is important. Attempts to provide better data onsisteny usually result in "resystematizing" or "onsolidating" existing appliations into larger ones with broader problem sope and data definitions. This may yield a satisfatory system within the defined sope, but again, as still broader problems are addressed, there will undoubtedly be data inonsistenies between the larger systems. Resystematizing at this sope may be extremely expensive and diffiult to justify, let alone aomplish. Comments prevail suh as "our systems annot talk to one another." What has been desribed is the lassial "bottom up" evolution of data proessing systems. n. order to begin to address the data onsisteny problem, a different philosophy must be adopted relative to data management. This is ommonly referred to as managing data as a resoure. This onept suggests that data is of onsiderable overall value to an organization and should be managed aordingly. t should be potentially available to and shared by the total business unit on a onsistent basis. t should not be ontrolled by a limited organizational segment but by a entral oordinator, muh like other orporate resoures, suh as ash and personnel. The management funtion would inlude formulating poliies and proedures for onsistent definition, souring, tehnial implementation, use, and seurity of the data. An ls Should Be Able to Live Through Organizational and Management Change Many data proessing systems and appliations are set up to provide the information needs of a speifi department or other organizational entity. Others are built solely on the speifi output report requirements of a partiular manager. Both types an beome immediately obsolete upon a reoganization or management hange. A new manager may have his own ideas as to what information is needed to run the department. While this kind of hange is inevitable, it an be expensive from a data proessing standpoint. The data proessing system, however, should in no way inhibit management flexibility in a dynami enterprise. Thus, the lis must antiipate and be apable of living through the long-term organizational and management hanges of a business with minimum impat if the expeted return on investments is to be realized. This objetive annot be realized without the proper support vehile for lis, and this vehile must be independent of the various omponents of the organizational struture. The BSP vehile is the business proess, that is, a basi ativity and deision area irrespetive of any reporting hierarhy or speifi management responsibility. A logial set of these proesses an be defined for any type of business and will undergo minimum hange as long as the produt or servie area of the business remains basially the same. One example of a business proess is purhasing. A partiular business might define this as "the proess by whih raw materials are aquired from vendors." There mayor may not be a separate organizational unit to ' aomplish this proess, or indeed there may be several. nherent within this proess are the various ativities and deisions neessary to aomplish the proess. Defining the organization's business proesses is one of the most important parts of the BSP methodology, and the method for doing so is tied diretly to the previously disussed l/s framework, that is, one based on resoores and planning and ontrol levels. With this in mind it is onvenient to define an organization's business proesses in assoiation with eah of its defined resoures. Emphasis in BSP is normally plaed 6

14 upon those proesses neessary to manage the key resoure. Eah resoure of a business an be thought of as having a "life yle" made up of several stages. A produt life yle, for example, has four stages: requirements,,aquisition, stewardship, and retirement. The time spread of the life yle an vary greatly with the partiular produt area but is of no onsequene in this approah. Business proesses an be identified to desribe the major ativities performed and deisions made by the business in the ourse of managing the resoure throughout its life yle. These an normally be organized into a proess hierarhy, and this is done without regard to organizational involvement or responsibility. The above approah results in proess definitions that enompass the three planning and ontrol levels previously disussed - namely, strategi planning, management ontrol, and operational ontrol. Using a produt resoure as the example again, the deision to pursue a partiular produt area would be "strategi planning," the planning and ontrol deisions relative to produt volumes or advertising expenditures would be "management ontrol," and the deisions in areas of engineering ontrol, manufaturing effiieny, et., would be "operational ontrol." By using this approah for all the resoures it is possible to define all the business proesses that take plae within any organizational segment. This may be tempered by pratiality in the atual BSP as there may be little /S support. interest for some of the resoures. The lis Strategy Should Be mplemented By Subsystem Within a Total nformation Arhiteture There are several impliations and onepts assoiated with this statement. The first is that a totall/s to support the entire business unit's needs is too big to build in any single projet. However, beause of the many problems assoiated with a "bottom up" evolution of systems (data inonsistenies, non-integrated system designs, expensive resystematizing, priority diffiulties, et.), it is very important that long-range goals and objetives for /S be established. The basi onept, then, is top-down lis planning with bottom-up implementation (Figure 3). With this onept the long-range /S goals and objetives are identified through a top-down planning proess (the BSP approah). The identified systems are then implemented in a modular building-blok fashion\ over time while remaining onsistent with the organization's business priorities, available funds, and other shorter-term onsiderations. This philosophy an be likened to the detail design and onstrution of a large offie building, whih would be unthinkable without an arhitet's approved drawing of the finished produt. Business objetives Business proesses Business organization.. Data proessing appliations Data files Data lasses Business objetives t Business proesses T nformation systems t Data bases t --- Planning Design and ----t1 implementation Figure 3" Top-down analysis with bottom-up it:nplementation The BSP methodology, although onsisting of onsiderably more steps and detail than shown in Figure 4, is onsistent with this philosophy. Stp of Figure 4, defining the business objetives, is intended to ensure agreement among all exeutive levels as to where the business is going, so that the lis strategy an be in diret support. Step 2, defining the business proesses, establishes the prime long-term basis for /S support in the business. Step 3, defining the data lasses, an be done on the basis of the proesses to be supported. A data lass, as the name implies, is a major ategory of data needed to support one or more business proesses. For example, ustomer information is a data lass needed in several proess areas, suh as order entry, billing, and distribution. This step results in a definition of all the data to be managed a&-3" resoure aross the business unit. Step 4, defining the information arhiteture, beomes a statement of the long-term lis objetive. This is normally in the form of a group of interrelated lis areas and the assoiated data to be managed. From the information arhiteture the individual modules an be identified, prioritized, and built as sheduled by the lis plan.. 7

15 Step 1 Step 2 Define the.. Define the business... business --- objetives proesses Step 3 Step 4 Define the Define the --.. information data lasses... arhiteture Figure 4. General /S planning approah A signifiant piee of the BSP methodology is the formulation of a reommended information arhiteture. The intent is to develop an implementation strategy that an be built in modules and yet provide justifiable return to the business at eah step. These modules or implement able piees are generally referred to as information systems (/S). Eah ould be thought of as the depository or management point'for a partiular set of the data lasses. As the data lasses are implemented (assumed through dta base tehnology), it is possible to provide the information needs for various business proesses. Eah /S then normally beomes assoiated with one or more business proesses and one or more data lasses (see Figure 5). The implementation strategy should obviously be in tune with the business needs. An identified /S is responsible for the olletion and maintenane of its data bases for the entire business. Other, or later implemented, systems an draw on these data bases as well as new ones that they will reate and maintain. Most of the identified /S areas will normally be in support of proesses that are operational in nature.. That is, the deision areas inherent in the proesses Wll relate to the "operational ontrol" level of planning and ontrol. Thus these /S's are sometimes referred to as operational ontrol systems. The maaged data is also at an operational level and is typified by muh detail and quantity. A management ontrol system, on the other hand (Figure 5), is in support of a "management ontrol" ategory of proesses. ts managed data would typially b summariztions of the internally generated operational data (dotted lines in the figure) plus any other data (ompetitive data, for example) as neessary to'support the proesses. By having "business plan" data in the data base, the. exeutive ould effet ontrol deisions by ompanng atual versus plan tor his area of responsibility. n summary, there are a number of basi onepts and philosophies relative to information systems upon whih the BSP methodology is based. The methodology itself should be onsidered flexible in nature. That is, ertain steps and tehniques ould be altered in order to adapt to speifi situations without detriment to the final outome. However, these basi onepts themselves should be onsidered inviolate. They, in effet, are BSP. 8

16 Produt Planning Distribution Manufaturing,, Management Control / / Marketing / / Administration Figure 5. nformation arhiteture example 9

17 Chapter 2. Overview of the BSP Study Approah The keys to suess in planning, developing, and implementing an information arhiteture that effetively supports the business goals are: Top-down planning with bottom-up implementation Managing data as a orporate resoure Orientation around business proesses Use of a proven, omprehensive methodology Chapter examined the first three at some length. This hapter gives an overview of a proven methodology for the BSP study whih will be delineated in more detail in the following hapters. Perspetive Sine BSP is part of an overall yle for providing the business with required information, it should be put into perspetive. As Figure 6 shows, there is a major junture between identifiation of overall business requirements and the six projet phases for implementing an js. Requirements are identified for a total business unit and then separated into projets that are undertaken and implemented over time. n addition to information systems projets, there is also a ontinuing set of projets overing information arhiteture and information systems management (SM). dentifiation of overall business requirements is not reiterative unless a major hange ours within the business unit that would hange the basi business proesses. BSP is a methodology for identifying the business requirements. f an entire business were not inluded in the BSP study, but only a omponent suh as one profit enter, additional BSP studies would be appropriate for other business omponents. n the past, and in many businesses today, projets are defined to address a funtional area of the business without regard to the total requirements of the business unit. BSP an provide overall diretion for the total business before projets are undertaken and therefore avoid the frationalization of data and inonsistenies of systems. Projet 1 (highest priority) V V V nformation Arhiteture Projets Projet 2 V V V Overall Business Requirements (SSP) Projet 3 V V V Projet 4 V V V Projet n V V V nformation Systems Management Projets time ljl Projet Phases (Optional). Requirements definition ". nternal design V. System test. External design V. Program development V. nstallation and maintenane Figure 6. Relationship of BSP to lis projets 10

18 The top-down, bottom-up aspets of the BSP study are refleted in the following: Business objetives Business objetives aining the. mmitment-.j r Preparing for the LstUdY Business proesses Business proesses Conduting the kikoff meeting Organization Defining business proesses Data proessing support nformation systems Data files Data lasses Analyzing businessl system re1ationships Data The study progresses from the very broad world in whih the business exists down to the data required to run the business. The data is ategorized into data lasses that lead to the definition of information systems to support the business proesses and business objetives. The study starts with a olletion of fats about the business that are usually available in doumented form throughout the organization. These fats are organized, abstrated, and analyzed by the study team and enhaned by the top exeutive, who explains the business and adds those points usually not doumented. The study progresses to an identifiation of the major ativities and deision proesses in the business and then eah of the exeutives is asked to validate and enlarge upon the fats that have been gathered and analyzed. The analysis ends with the onsolidation and omparison of the fats from all soures. From this understanding the study follows the normal path of findings and onlusions, reommendations, ation plan, and exeutive presentation for onurrene and support. Major Ativities As Figure 7 indiates, there are two major ativities that preede a BSP study and eleven in the study itself. Determining the exeutive perspetive' Assessing business problems and benefits Developing reommendations and ation plan Figure 7. Flow of the BSP study 11

19 While these ativities may be arried out in varying degrees, none an be omitted. Figure 7 shows these ativities and their most logial arrangement. When one beomes fairly familiar with the BSP approah, however, he an, as appropriate, do part of a given ativity and delay the balane. The remainder of this hapter is a brief ommentary on these major ativities. Gaining Commitment A BSP study should not be started unless a top exeutive sponsor and some other exeutives are ommitted to being involved in it. The study must reflet their view of the business, and the suess of the study depends upon their providing the business understanding and information requirements to the team. Most of the input will ome,diretly or indiretly from these exeutives. Sine approval of study reommendations ommits th ompany for several years to a ertain diretion in the use of its data proessing resoures, it is important at the outset to get agreement on the sope and objetives of the study and on its expeted deliverables, so as to minimize future misunderstandings. The most important ation following ommitment regards seletion of the team leader, an exeutive who will work full time in the study and diret team ativities. He sees that ontat with other exeutives is on the proper level and that input from them is interpreted orretly. A letter from the sponsor to a,ll partiipating exeutives sets the tone and signifies ommitment. Preparing for the Study Considerable saving of time, avoidane of frustation, and higher quality of output an be gained by proper study preparation. All exeutive partiipants and the team need to know what will be done, why, and what is expeted of them. Proper eduation and orientation will provide the best input from the exeutives and the best use of it by the team. nterviewees are seleted as soon as possible so as to allow for their orientation, sheduling of interviews, and the providing of information to the study team. F or maximum effiieny on the part of the team in working together full time during the study, information on the ompany and on data proessing support is gathered before the study kikoff. A ontrol room is established so that the team may work together, display relevant material on the walls, and ondut interviews. The major output should be a study ontrol book ontaining: a study work plan; a shedule of interviews and a shedule of hekpoint reviews with the sponsor; an outline of the final report from the BSP study; and business and information systems data, analyzed and harted, and ready for the kikoff period. This ativity should end with a review by the BSP study sponsor. Conduting the Kikoff Meeting The BSP study itself and the full-time partiipation of the team members starts with the kikoff meeting, whih onsists of three presentations. First, the exeutive sponsor reiterates the objetives, expeted outputs, and perspetive of the study with relation to other ompany ativities and objetives. The next presentation is onerned with the main purpose of the kikoff, whih is to provide that eah team member is onversant with the information that has been gathered and to disuss those fats that are not part of the information supplied. To aomplish this the team leader "walks through" the business fats that have been gathered and makes subjetive omments and additions on fats that annot be readily doumented - politis and sensitive issues, and hanges planned and in proess He should also over the deision proess, how the organization funtions, key people, major problems, the user's view of DP support, and the image of the DP department. The third presentation is made by the information systems diretor or one of his managers, who gives the team a view of data proessing analogous to what was presented for the business. He should also over projet status and projet ontrol, history of major data proessing projets started in the last two years, major urrent ativities, planned hanges, and major problems. These three presentations, added to the fats that have been gathered and made readily available to the team, should give the team an overall understanding of the business and the present and planned data proessing support. Defining Business Proesses No other ativity during the study an be quite as overwhelming or as important as the identifying of the business proesses. Sine these proesses form the basis for exeutive interviews, the information arhiteture, problem analysis, data lass identifiation, and various follow-on ativities, everyone on the team must aquire an understading of all the proesses, and they an do so by assisting full time in their identifiation and in the. writing of their desriptions. The major output from this step will be a list of all the proesses, a desription of eah, and the identifiation of those that are key to the suess of the business. Defining Data Classes The defining of data lasses is the grouping of data into logially related ategories. This lassifiation, and its 12