GISdevelopment.net ---> GITA 1997 ---> Project planning, implementation and management

Cost effective AM/FM/GIS CAD conversions,on time

Jeff Head
Mid-StatetinSUki311tS,hlC.
1500 North, 200 West, Nephi, UT 84648


Setting realizable goals
The conversion of manual graphic data into CADformats of one type or another is a processthat has been evolvingfor almost 20 years. Manyfirms have gonethrough severalgenerationsof CADsofbvarein an attempt to realize gains from automationwhich haveproven elusivefor many. An analysisof howthese systemshavebeen soldand purchased will showthat in too manycases,due to budgetaryor scheduleconcerns,decisionshavebeen basedon minimal up-front analysisand the demonstrationsand claims of the vendorsand suppliers,and not on a structuredanalyticalmethod. With advents in technologyprovidingprocessingand memoryat lowprices, and great strides in data base software,the potentialfor truly realizing cost effectiveautomationgains has arrived. Byemployingstructuredanalysis methodologies,the claims of vendorsand supplierscan be filteredto find the successfulsolutionfor virtually any firm.
  1. Critical nature of defining business and technical goals up-front
    There are rnauycritical issuesto be consideredwhen developinga structuredplan for any major conversionprocess. The foremostof these are the businessand technical goals to be achievedas a result of the conversion. These must be carefully thought out with regard to their impact to the firm as a whole,and they must be considered well before conversion begins. They must also be thoroughly documented so they can guide all phases of the conversion project.


  2. Business Goals
    While the graphic data from the engineering and design areas of any firm dealing in AIWFM/GIS could be viewed as purely a technical issue, those with a broad view know that the potential for integrating significant poxtions of this data with business data tied to the same features is significant. Issues from cost analysis, to inventory, to billing, to dispatching and assignments, and many more, can be positively impacted if the design portion of the firm and the business (accounting,billing, purchasing,etc.) portion of the firm are keying off of common data. Through involving other areas of the company, greater buy-in and greater positive impact can also be achieved. In order to accomplish this, a thorough analysis of the issues and their impact on the business process and communication channels within the firm, and the cost of achieving them, should be completed well in advance of actual data conversion.


  3. Technical Goals
    Technical issues are historically the focus of conversion projects. While this is critical to the engineering, design and drafting functions, technical goals alone will provide only local, discipline specific gains if not considered in light of the business considerations outlined above. These local gains can be signiticanly and require careful thought and consideration well in advance of the conversion, and range from allowing for more accuracy, to quicker turn around of change, to less change, to better standards, to better maintainability and archival, and to any numberof other automation benefits. When considering the impact desired, it is important to consider the effect on disciplines and groups that will interface with the data and how they will communicate with one another after the change. As has been stated, this can extend into business functions where the impact can bejust as significant.


  4. The Key to Realizable Goal Settinng for Conversion Proiects
    It must be emphasized, significant automation will impact the way you do business. It will impact how various groups and disciplines communicate and interface with one another. Unless clear goals are developed in this light significant “ripple effect” can occur between groups and disciplines. The cost of communication and interface “ripples” in an organization can off-setor negate the expected gains of automation. Without this Perspactive, and the investment required to develop it, realizable goals become the exception rather than the rule.
Technology considerations
There are many technology considerations to be considered when contemplating CAD conversion. These include all aspects of several main areas of focus. These can include:
  1. Data Base Considerationsfor any integration.
  2. NetworkingIssues.
  3. The CADsysteminto which the data is to be converted.
  4. Conversiontechnologiesand methodologiesavailable.
All of these must be consideredas business and technical goals are established to guide the overall process, This is a critical part of the structured approach to developing an overall plan for conversion.

  1. Data Base considerations
    When developing an AM/FM/GIS land base tied to the business and engineering functions of any firrm the database considerations take on tremendous importance. Should there be a central data base, should the data base be distributed or networked should duplicate records be maintained, how are duplicate systems reconciled and at what COst how large can a central or distributeddatabase become,howare the databases segment@ what accessprivilegesare requiredfor difTerentgroups,what type of networkis requiredfor the users, etc.. All of these questions,and more, are extremelycritical. Current technologies,either alreadyin place at the firm, or to be acquiredas a result of the project must be analyzedand their capabilitiesmeasuredagainst the answersto these questionsand the goals for the project. Generally,the less duplicationand needfor reconciliationbetweendata bases, the better. Resultinganswersand plans must be consideredin light of their impact on cmmmmicationchannelsand interfacesbetweengroupshiisciplines.


  2. The target CAD System
    CADsystemshavebeen evolvingsincethe late 1970’s.Selectinga CADsystemhas been a task which many companieshavegone through severaltimes. Generally,everythingthat is being stated here about a stnwtmd analysisapproachto conversio%can be applied to the selectionof a CADsystemas well. In the case of conversion projectsseekingto take full advantageof any AIWFIWGISmnversiou existing CADsoftwaremust be analyzed against the business and technical goalsbeing establishedfor the conversion. If the goals establishedfor the project cannotbe achievedwith existing software,then the cost of new softwarethat accommodatesthe goals needsto be measuredagainst the levelof attainment possiblewith the existing software. If new sofhvareis required,then it shouldbe purchasedthrough a structuredanalysismethodthat uses the goals established for the conversion project and the general requirements for design and drafting, to developa specification. From there, a pilot, or benchmark, should be establishedto evaluate vendors. Do not rely on goodmarketingpresentations,or “wow”demos.


  3. Conversion Technologies/Methodologies available
    There are severalmethodologieswhich havebeen developedand are employedto convertmanual drawingsand data to digital format. It is critical to selecta method after careiidlymeasuringthe various options against the goalswhich havebeen establishedfor the conversionproject. In additioma numberof basic questionsneedto be considered beforeselectinga method. Suchquestionsas: Are the resulting files mainlyfor archivalpurposes,or will they be activein the desigrdengineeringprocess? Is there associateddatabase informationattachedto the graphicsfrom whichtechnical or businessfunctionskeyoff of? What degreeof accuracyand tolerancingis required? Howmany layersor levelsare required?What standardsand conventionsare employed?Howlarge are the drawings?Howdense are the graphics?What is the state of the existing media (copies,blue line, micro-film etc.)?Thesetypes of questions, and issues regarding relativecosts of the various options, will have an impacton the type of methodologyemployedto convertthe drawings. Table 1 showsvarious conversionoptions availableand their pro’s and con’s.

    Table 1- Basic Conversion Methodology Comparison


    The above table is meant to be a basic comparison of methods. Depending on the goals established,any of the methodscan be applied successfully,if the goals and the inherent strengths and weaknesses are fully considered. Many times a combination of methods may be required. In any case, it is critical to develop a thorough specification based on the business and technical goals, the technical requirements(standards,conventions,D/B design, etc.), and the budgetary and scheduling issues. Such a document will serve as a guide through the entire process.
Developing a thorough specification
Once the business and technical goals for the conversion project are established,and once the technical considerations involved with achieving those goals have been thoroughly analyzed,a specification which defines the requirements for the project can be developed. This document becomes the detining document for the strwhmd processutilized to completethe job. The specificationshould defineall of the parametersand establishmeasurementcriteria. The developmentof the specificationshouldbe accomplishedby a team which consists of individualswho havethe technical and/or businessbackgroundsto insure that all of the goals and technical issues are taken into considerationand adequatelycoveredby the specification. This is a task which shouldbe adequatelybudgetedin terms of time, personnel and dollars. Dependingon the amount of detail, the scopeof the job and the amount of technical sophistication required,this task could take anywhereffom severaldays,to a fewweeks,to two or more months. The time invested will be wellworth it becausethe successof this processwill makethe differencein attaining the goals that havebeen established. A sampleoutline for such a specification,with a fw of the possiblesubheadings,follows:

Sample Outline for a ConversionProiect Mecification
  1. Introduction
    1. BackgroundInformation
    2. Basic Descriptionof Business/Work
  2. Scope of Work
  3. ConversionMethodology
    1. Methodology Description
    2. Software Requirements
  4. Existing Data Type
  5. Target CAE/CAD System
    1. Description of System
    2. Expertise/Training Required
  6. MeasurementUnits & Coordinate System
  7. Tolerance Requirements
  8. Conventions
    1. Naming
    2. Layering
    3. Color
    4. Speciiic
    5. Etc.
  9. Standards
    1. Scale
    2. Syrnbdogy
    3. Dimensions
    4. Text
    5. Drafting/Design(MIL, 1S0, etc. Specs)
    6. Etc.
  10. IntegratedData BaseRequirements
    1. Structure
    2. Sources/Input
    3. Verification
  11. Schedule Requirements
  12. Conversion Guidelines
    1. Scrubbing Requirements
    2. Set Up and Review
    3. Delivery Format
  13. Quality Plan (See Sectionof this Paper, “Developing a ThoroughQualityPlan”)
  14. DocumentTracking & Reports Attachments (Including MIL or 1S0 Specs, Symbology, Software Listings, Menus, etc.)
The aboveoutline is a sample. There are manyother headings, subheadings,and categoriespossible.The key is to establisha specificationthat reflectsin full the businessand technical goals of the project,the technical considerations requiredto obtain those goals, and all of the standards,conventions,timeframes and methodologiesthat are required. In addition, the definitionof the qualilyplan and the measurementcriteria is critical. This establishesthe standardby whichthe completedwork is m- and through which any reworkrequirementsare determined.

Developing a thorough quality assurance plan
The qualityplan shouldbe developedin a Total QualityManagement(TQM)environment. This meansthat managementmust fidly supportthe stmctmd methodfor developingthe projectfrom the onset, and in particular the resourcesrequiredto i- a qualityoutcome. It also meansthat the entire processshouldbe gearedtowardsinsuring that quality is of prime considerationthrough everyphase of the project. This includesall aspectspreviouslydiscuss@ the establishmentof realizablegoals, the analysisof the technical considerationsand the developmentof the specification,which includesthe qualilyplan. It also extendsinto the later phases of the project such as the developmentof a pilot, the selectionof seMce suppliersand the acturdimplementationof the conversionproject followedby the implementationof the finished product. Keyto the qualityplan shouldbe a thorough understandingby all parties of all expectations,measurementcriteria and analyticalmethods,statisticalmodeling,and actual verification processesthat will be employed, Also, the consequencesof sub-paror exceptionalwork shouldbe described. With this in mind, a sampleoutline of a quality plan follows:

Sample Outline for a Conversion Proiect Ouality Plan
  1. Scope of Quality Work
    1. Items to be Checked
    2. Accuracy Indicators
  2. Methodology
    1. Manual Components of Check
    2. Automated Components of Check
  3. Statistical Model
    1. Statistical Method
    2. Model Definition
  4. Target Quality Measurement
    1. Baseline Data Definition
    2. Maintenance and Verification of Baseline
    3. Accuracy Required
    4. Incentives
    5. Penalties
  5. Re-Work
    1. Criteria
    2. Time Frames
Once defined the quality plan becomesthe driving forcebehind verificationof the productiondata. It cannotbe overemphasizedhowimportant this documentis, and howwell thought out it shouldbe. If the specificationis the foundationfor the project, the qualityplan becomesthe cornerstone. Howwell the quality plans definesthe critical componentsof the project(uponwhich the goals rest), howwell it is understoodand howwell it is adheredto will determineto what degreethe goals are accomplished,and what degreeof confidencecan be had in the outcome.

Evaluation solutions
With the goals for the projectestablished,a specificationdefining all of the requirementsin place and a quality plan developedgoverninghowthe finished product is to be verified,a critical evaluationof the various solutionsavailableis possible. This evaluationshould include a look at internal and external resourcesavailableto completethe process. In most cases,forjobs that are relativelylarge requiring a fairly aggressiveschedule,outsideresourcesare employedto completethe bulk of the work, while internal resourcesare marshaledto ver@ the work. In order to evaluatethe various options, severalstmctmd stepscan be taken whichwill allowfor a determinationof which option(s)best meet the overall requirementsin a quality fashion,and thus maximizethe attainment of the goals. These steps includethe establishmentof a bid processwith overall requirementsfor the bidders, the establishmentof a pilot projectand its implementationagainst a selectgroupof bidders, and the awardingof the conversioncontractto one or more suppliers basedon performanceduring the pilot.
  1. Establishing a Bid Process
    A Requestfor Bid (RFB)can be establishedthat definesthe methodologyto be employedduring the evaluation process. The RFBis a critical documentbecauseit establishesminimumcriteria for bidders, introducesselected goals, introducesthe specificationand quality plans and detailshowthe pilot projectwill be employedfor a select groupof bidders. In this respect,the bid processbecomesa two phased approach. The first phase establishes minimum criteria against which all bidders are gaugedto determinea groupof bidderswho will then be piloted. Someof the criteria that should be considered in this initial phase include:

    1. The financial conditionof the bidder
    2. The technical historyand expertiseof the bidder
    3. Referencesfrom the bidder that directlyrelate to the type of work on this project
    4. The size of the bidder, or the bidders sub-contractors(Crmthe bidder supportthe project)
    5. The methodologyemployedby the bidder (Howwouldthe bidder approachyour project)
    6. A samplethat the bidder can completeand make initial cost estimatesagainst

    A requestfor this informationshouldbe presentedto bidders either in a bidders’conferenceor by mail with specific deadlinesfor the receptionof the information. The resulting informationshouldthen be evaluatedagainst a set of criteria alreadyestablishedand weightedfor this purpose. Dependingon the size and complexityof the overall proj~ a final set of bidders (short list) is then establishedfor the pilot phase of the evaluation.Anybidderwho qualifiesfor the pilot phase shouldbe viewedas fulfillingthe initial criteria and capablein that regard. Oncethe list of bidders is shortened the pilot phase of the evaluationcan proceed.


  2. Establishing and Implementing a Pilot
    A pilot projectshould fulfill all of the followingcriteria for the companyemployingit:

    1. Represent all major business and technical criteria.
    2. Be verifiable per the quality plan against alreadycompletedresults.
    3. Be conciseand accomplishedwithin a manageableevaluationschedule.
    4. Be funded.
    5. Have a set of evaluation and weighting criteria already established

    . Bidderswho have qualifiedfor the pilot phase should receivea packagethat details the pilot processand includesthe completespecificationand qualityplan. This packageshouldalso give a detaileddescriptionof the pilot project itself along with the time framesand schedulesassociatedwith the pilot. In somecases,overall cost estimatesfor the productionworkmayultimatelybe basedon the pilot. In this case, criteria and proceduresgoverninghowthe bidders bid the overallprojectand howtheir financialbids are evaluatedmust be included. It shouldbe emphasizedto the bidders that the quality plan is of utmost importanceand will be a keyfimctionin assisting with the final evaluation. Dependingon the dispositionof the weighting methodologies,the manner in which daa costs,timeframes, goal accomplishment etc. are weightedand evaluatedmaybe sharedwith bidders. Generally,weighting methodologies are kept confidentialso that bidderswill attempt to completethe entire pilot in a qualitymanner.

    Once the pilot is implemented,it becomesa soundingboard not only for the bidders,but for the specificationand productionprocess,including quality assurance. This shouldbe taken into accountas bidders are evaluated. Chances are, that the processwill be modifiedas it is pilote4 thus requiring somemodificationin evaluationcriteria as the pilot progresses. The result of a pilot carried out in this manner is a product that validates goals, specificationsand bidders.


  3. Awarding the conversion contract
    The awardingof the conversioncontract shouldbe basedon the weightedcriteria establishedfor the pilot, and the evaluationof this criteria against performanceon the pilot. This results in an empiricalmethodwhich can clearly indicatethe order of performanceagainst the pilot. Non-empiricalissues (whichwerethemselvesevaluatedagainst weightedcriteria to producebidderswhocouldparticipatein the pilot phase)werealreadycoveredin the initial phase of the bid processdescribedearlier. A sampletable showingweightedcriteria evaluatedagainst a set of bidders is shownin Table 2.

    Table 2- SampleEvaluationMatrix based on Weighting Criteria

    The above table employsweightedcriteria evaluatedagainst fivevendors. There are two weighting methodologies employed. The first is an incident countwhich countseach incident abovethe target qualitymeasurementin specific areas negatively. The secondare positivefactorswhich are weighted,evaluatedand then subtractedhorn the incident total. In the abovetable, if twovendorswere selectedto be awardedproductionconversioncontracts,VendorE and VendorB wouldbe awardedthe contracts. If the contractwasto be awardedproportionally,then VendorE wouldbe awardedapproximatelytwo-thirds of the contractand VendorB wouldbe awardedapproximatelyone-third.

    In a similar t%shion,using the abovetable, if three vendorswereto be awardedthe project,VendorsE, B and A wouldwin. The proportionateportions of the contractwouldfind, VendorE at 58’%0, VendorB at 30°Aand VendorA at 12°Aof the total contract.
Completing the conversion project
Oncethe productioncontractsare awarded,the projectcan proceedutilizing the specificationto control the production and the qualityplan to measureresults. The productionschedule,outlined in the specificationand as stated earlier in this paper, should provideadequateopportunityto continuouslyapply the results of the quality plan as the project proceeds, In other words,there shouldbe enoughsegmentsor divisionsof the productionwork to allowfor the results of the qualitychecksto be applied against remainingportions of the projectas the scheduleproceeds. This is a key elementof the TQMprocess. If the earlier phases of the projecthavebeen well evaluat~ and implementedeffectively, the productionwork shouldproceedsmoothly,as long as continuousimprovementhas been allowedfor and is applied. All parties in the process(both internal to the companydesiring the conversion and the external contractorscompleting variousportions of the work)must understandthis going in.

The wisdom of awarding multiple contracton largerjobs will also help insure timely completionof the work. Each vendor shouldhave the capacityof completingmoreof the workthan they are awarded. In this manner, if some unforeseenpolitical disaster, natural disaster or other “Actof God”wereto disrupt one contractor,contingencyplans wouldallowfor transfer of work to another. Conclusionsregarding geographicaland international locationsof contractorsin this regard are clear, there shouldbe someseparation. The old adage of not “putting all of the eggs in one basket”can be applied.

Summary
In summary,a stmctured analyticalmethodcan be appliedto AM/FM/GISconversionprojects. This methodcan establishrealizablegoals, developthorough specitlcationsand qualityplans, allowfor effectiveevaluationand piloting of contractor and lead to effectivecompletionof the project,on time. Althoughrequiring moreupfkont costs, the structuredmethodsdescribedin this paper can lead to the tremendouscost and productivityimprovementsthat have long been envisionedby automationusing CAD. The gains from such successwill causethe initial cost requiredto insure them to pale in comparison.
© GISdevelopment.net. All rights reserved.