Software Project Planning After the finalization of SRS, we would like to Software ee taaieieaes Project Planning finalization of the SRS. Software Project Planning Software Project Planning In order to conduct a successful software project, we the software evolves from concep to realty, and culminates only must understand: en the software fe rete = Scope of work 1 be done = Therrisk o be incurred The resources required = The task to be accomplished = The cost to be expended Fig. [Activities during Software Project Planning The schedule to be followedSoftware Project Planning Size Estimation Fig, 2: Function for sorting an array Lines of Code (LOC) If LOC is simply a count of the umber of lines then figure shown below contains IsLoc When comments and blank lines are ignored, the program in figure 2 shown below contains 17 LOC. Software Project Planning Software Project Planning Growth of Lines of Code (LOC) Software Project Planning Furthermore, if the main interest is the size of the program for specific functionality, it may be reasonable to include executable statements. The only executable statements in figure shown above are in lines 5-17 leading to a count of 13. The differences in the counts are 18 t0 17 10 13. One ccan easily see the potential for major discrepancies for large programs with many comments or programs written in language that allow a large number of descriptive but rnon-executable statement. Conte has defined lines of code “A line of code is any line of program text that is not a comment or blank line, regardless of the number of statements or fragments of statements on the line. This specifically includes all lines containing program header, declaration, and executable and —-non-executable statements". This is the predominant definition for lines of code used by researchers. By this definition, figure shown above has 17 LOC.Software Project Planning Function Count Alan Albrecht while working for IBM, recognized the problem in size measurement in the 1970s, and developed a technique (which he called Function Point Analysis), which appeared t0 be a solution to the size ‘measurement problem, Software Project Planning “The FPA functional units are shown in figure given below: © Fig. 3: FPAs functional units System Software Project Planning ‘The principle of Albrecht's function point analysis (FPA) is that a system is decomposed into functional units, puts information entering the system = ups information leaving the system = Enguises requests fr instant acess 0 1 oxen logical les information hel within the system stem interface files information held by ote systema thats sed bythe syst blag analyzed Software Project Planning ‘The five functional units are divided in two categories: (i) Data function types = Internal Logical Files (ILF): A user identifiable group of | logical related data or cootsol information maintained within the system, * Extemal Interface files (EIF): A user identifiable group of logically related data or control information referenced by the system, but maintained within another system. This ‘means that EIF counted for one system, may be an ILF in another system,Software Project Planning (Gi) Transactional function types ‘External Input (ED: An EI processes data or contol information that comes from outside the system. The EI is an elementary process, which isthe smallest unit of activity that is meaningful to the end user inthe business, 1» External Output (EO): An EO is an elementary process that generate data-of control information to be sent outside the ‘= External Inquiry (EQ): An EQ is an elementary process that is ‘made up 10 an input-output combination that results in data retrieval. Software Project Planning ‘> Function points are directly linked to the statement of requirements; any change of requirements can easily be followed by a re-estimat. > Function points are based on the system user's external view of the system, non-technical users of the software system have a better understanding of hat function points are measuring. Software Project Planning ‘Special features ‘> Function point approach is independent of the language, tools, oF methodologies used for implementation; Le. they ‘do not take into consideration programming languages, ‘data base management systems, processing hardware or any other data base technology. > Function points can be estimated from requisement specification or de specification, thus makin possible to estimate development efforts in early phases of development. Software Project Planning Counting function points, [External Inputs (EI) 3 4 6 [Extemal Output (EO) 4 5 7 [Extemal Inquiries (EQ) 3 4 6 [External logical les (LF) _| 7 10 |_15 [Extemal Interface fles (EF)] 5 7 10 ‘Table 1: Functional units with weighting factorsSoftware Project Planning LOCO NOO Oooo LOOCOUWOonoo noo oo o 0 Software Project Planning Software Project Planning ‘The weighting factors are identified for all functional units and multiplied with the functional units accordingly. The procedure for the calculation of Unadjusted Funetion Point (UFP) is given in table shown above. Software Project Planning ‘The procedure for the calculation of UFP in mathematical form is given below. Where indicate the row and j indicates the column of Table 1 W, Tes the cry ofthe row adj colum ofthe able 2: is the count of the numer of functional units of Type # that have heen classified as having the complexity comesponding 10 colina ‘Organizations that use function point methods develop aeriterion for determining whether a particular entry is Low, Average or High Nonetheless, the determination of complexity is somewhat subjective A UFP* CAP Where CAF is complexity adjustment factor and is equal to [0.65 + O01 x SF). The F, (21 wo 14) ae the degree of influence and are ‘based on responses to questions noted in table 3Software Project Planning "0 tte omalpmsing ots 1 oar aetgat tcc? 12 sonnel cl te sg’ 1 gaa Software Project Planning Example: 4.1 Consider a project with the following functional unit: [Number of user inputs =50 [Number of user outputs 40 Number of user enquiries 35 Number of user fi [Number of extemal interfaces 08 Assume all complexity adjustment factors and weighting factors are verge. Compulte the function points for the project. Software Project Planning Functions points may compute the following important metres: Productivity FP /persons-moaths Quality Defects FP Cost Rupees / FP Pages of documentation per FP Documentation ‘These metres are controversial and are not universally acceptable ‘There are standards issued by the International Functions Point User Group (IFPUG, covering the Albrecht method) and the United Kingdom Function Point User Group (UFPGU, covering the MKIL method). An ISO standard for function point method is also being developed. UBP =50x4440x5435x4+6x 1044 x7 200 + 200 + 140 + 60+ 28 = 628 (0.65 + 0.01 SF) = (0.65 + 0.01 (14x 3)) = 0.65 + 0.42 = FP =UFPxCAF 628 x 1.07 = 672 CAFSoftware Project Planning Example:s.2 ‘An application has the following 10 low external inputs, 12 high external outputs, 20 low internal logical files, 15 high external interface files, 12 average external inquiries, and a value of complexity adjustment factor of 110. What are the unadjusted and adjusted function point counts ? Software Project Planning TIEEES CConsidera projet with the following parameters (External inputs (2) 10 with tow compeity (6)15 with average complexity ()17 with high complexity (Gi) External Outs (26 with low complexity (6)13 with high complexity (iy External Inquiries: (@) 3 with low complexity (6) 4 with average complexity ()2 high complexity Software Project Planning Solution Unadjusted function point counts may be calculated using “ 33 UFP=YYZ,w, 0x34 1247420074154 1041284 2048441404 1504.48 482 Fe ZURPxCAF 452% 11 Software Project Planning (Go) Invernal logical Files: ()2 with average complexity (6) with high complexity (6) External Interface files: (4)9 with low complexity In ation to above, sytem requires nificant data communication Performance is very critical Designed code may he moderately reusable iv. System is not designed for multiple installation in different ‘ofganizatons. ‘Other complexity adjustment factors are treated as average. Compute the function points forthe project.Software Project Planning ‘Solutions Vaated fncon pons ay Besa ng le 2 i SS apa SS ost So com : (eae Ss feet et oes ed = = Software Project Planning Folate Gost of Software Phases Software Project Planning SIF, HBSS ISSO CAF = (0.65 +0.01 x EF) (0.65 + 0.01 x 41) Software Project Planning Cost to Detect and Fix Faults etctane caret ftSoftware Project Planning Cost Estimation |A number of estimation techniques have been developed and are having following atributes in common > Project scope mustbeestalshedin advance Software metics ofa used as a bai om which estimates are made 1» The ojectis broken ino small pcos which re estimated nvidully To achieve reliale cos and schedule estimates, a number of options ‘> Delay estimation untae in project 7 Use simple decompositon techniques to genarate project cost and ‘theduaeetatee > Develop empirical medals or estimation 1» Acquire one or more automated estimation tols Software Project Planning Software Project Planning MODELS Static, Single Static, Variable Multivariable Models Models Static, Single Ve jodels “Methods using this model use an equation to estimate the desired ‘values such as cost, time effort, etc. They all depend on the same variable used as predictor (say, size). An example of the most, ‘common equations is Coal (Cis the cost, Lis the size and bare constants EB = 14 DOC = 30.4 Lo D =46L0% Effort (E in Person-months), documentation (DOC, in number of pages) and duration (D, in months) are calculated from the number lines of code (Lin thousands of lines) used asa predictor. Models often based on equation (i), they actualy depend con sever variables representing various aspects of the software development envionment. for example method used, user participation, customer oriented changes, memory constants, BE =52L0 D =41 Le ‘The productivity index uses 29 variables which are found to be ted to productivity as follows:Software Project Planning Example: 44 ‘Compare the Walston-Felix model with the SEL model on a soltware development expected to involve 8 person-years of effort, (@)Caleuate the number of lines of source code that can be produced, (Calculate the duration ofthe development, (Calculate the productivity in LOCIPY (Calculate the average manning Software Project Planning (@) Darton in moms can be calculated by meas of equation DiSEL) = 4.649% = 4.6 (08264) DOW.) =A LO 3 months = 4.1024.632)" () Productivity isthe lines of code produced per person/month (year) «sex = 284. 17830 Person ~ Years PW-P) 3079 L0C | Person Years Software Project Planning PY = 96 person-months (a) Number of tines of source code can be obtained by reversing ‘equation to give: Lea 96/1 40% = 94264 LOC (9605 2)! = 24632 LOC, Software Project Planning (4) Average manning i the average numberof persons required per ‘month inthe project. (SEL) = oe 6. sPersons 7 = 55P=M 74 Persons 13MSoftware Project Planning ‘The Constructive Cost Model (COCOMO) Constructive Cost model (COcoMO) Basic Intermediate Detailed “Model proposed by BW, Boehm's through his book Software Engineering Economics in 1981 Software Project Planning Software Project Planning COCOMO applied to Semidetached Organic “mode ‘Embedded mode mode Software Project Planning Basie Model Basic COCOMO model takes the frm E=a,(KLOC)* D=c,(E)" Where E is effort applied in Person-Months, and D is the development time in months. The coefficients ay. by cand ch afe siven in table 4 (a,Software Project Planning Sotiare a a | & | & manic ze | sos [2s | oa fsemisacnes | 30 | tx | 2s | os eroedsns as | am | 2s | om “Table a): Sais COCOMO cateets Software Project Planning Software Project Planning ‘When effort and development ime are known, the average staff size to complete the project may be calculated as: Average te (SS) = Persons When project size is known, the productivity level may be calculated as Software Project Planning Example: 45 Suppose that a project was estimated 1o be 400 KLOC. Calculate the effort and development time for each of the three modes i., organi, semidetached and embedded, Solution The basic COCOMO equation take the form: E=a,(KLOC)* D=6,(KLOC)" Enimated size ofthe project Organic mode 41400)! 2.512951)"Software Project Planning Software Project Planning 5) Semidetached mode Example: 46 = 30(400)!"#= 2462.79 PM. A project size of 200 KLOC is to be developed, Software D=25(2462.79)035238.45 PM development team has average experience on similar type of projects, The project schedule is not very tight. Calculate the effort i) Embedded mode development time, average staf size and productivity of the project. = 3.61400)" b. (77281 PM 25(4772.8)9= 38 PM. Software Project Planning Software Project Planning “The semi-detached mode isthe most pproprite mode; keeping in view the size, schedule and experience ofthe development tam, 0200) 12= 1133.12 PM D=25(1133.12)=203 PM Productivity =KLOC __200_ E 1ishi2 Hence 1765 KLOC/ PM P 76 LOC! PM Average sta size ( = Persons D M3812 47 Persone‘oftware Project Planning Software Project Planning Intermediate Model (i) Personal Atsibutes pal 1 Programmer capability Cost din (Product Atriba 1» Required sh reliability » Analyst 1p Size of application database ~ Application experience > Virtual mfc experience > Complexity of the product 1 Programmin (Gi) Hardware Anbu ‘> Run time performance constraints programming practices > Memory constraints 1 Use of software tots 1 Viral machine volatility 1 Requited development Schedule 1 Tumaround time Software Project Planning Software Project Planning “Mulplirs of different cost drivSoftware Project Planning Intermediate COCOMO equations 6 (Ey a,(KLOC)" * EAF Software Project Planning Detalled COCOMO Mode! Detsileg, COCOMO teste ee | Phase-Sensitve Three level product effort multiplies hierarchy omarie | a2 | 10s | 25 | 038 ome | 2 fe | es con Modules subsystem emitac drivers design ‘System level Embedded 28 1.20 25 0.32 & west Manpower allocation for Tee conan seco ; veer Software Project Planning Software Project Planning Development Phase Design Plan Requirements Bat: 168 oi EFFORT 6% 10 8% 4 Programming DEVELOPMENT TIME : 10% to 40% fon % depend on mode & size ime AE oot Integration & Test fon tow 348 rine teesSoftware Project Planning Software Project Planning Principle of the effort estimate Size equivalent [As the software might be partly developed from software already ‘existing (that is, re-usable code}, a fll development is not always required. In such cases, the pars of design document (DD), code (Coe) and integration (Se) to be modified are estimated. Then, an adjustment factor, A, is calculated by means of the following ‘equation. 4DD+03C+031 The size equivaleais obtained by (equivalent) =(S x A)/ 100 E,=u,E D,=1,D Software Project Planning iecycle Phase Values of, Lifecycle Phase Values of {1, ToaICoR | Fant] Spm oa asa or ste [mecaomns | Socom [Seam [coset ten | "Ee Software Project Planning Distribution of software life cycle: 1. Requirement and product design (@)Plans and requirements (System desi Detailed Design (Detailed design 4. Code & Unittest (@)Module code & test 4. Invegrate and Test (aplmtegrate & Test,Software Project Planning Example: 47 [A now project with estimated 400 KLOC embeded system has 9 be ‘developed. Project manager has a choice of hirag fom two pools af ‘developers: Very highly capable with very litle experience in the Programming language being used + Developers of low quality but alot of experience withthe programing language. Whit i the impact of hiring all developers from ne or the other pool? Software Project Planning (Case I: Developers are of low quality but lot of experonce withthe programming language being used, EAP =1.29%0.95-1.22 E 712 12 > 5 (4528)°% Case I requires more effort and ime. Hence, low quality developers wh lot of programming language experience could not match with the pertormance of very highly capable developers wih very liter experience, Software Project Planning Solution ‘This is the case of embedded mode and model is intermediate cocoma, Hence E=a,(KLOC) . (400) 712 PM Case I: Developers are very highly capable with very ite experience in the programming being used EAF =082x 1.14= 0.9948 e 712 x 9948 = 2470 PM o 5 (3470) =339M Software Project Planning Bample: 48 Consider «projet to develop fll seeen editor. The major components ‘deni 1. Screen ett 1. Command Language Intrpeter Fite Inpat & Output 1V.Cursor Movement V. Screen Movement “The size ofthese ae estimated wo be dk, 2, Hh, 2k and 3k delivered source Codeine, Use COCOMO to deters 1. Ovenll cost and schedule estimates (assume values fo ferent ‘ost divers, with at lest thee of them being dfleent rm 1.0) 2. Cou & Schedule estinutes for diferent phasesSoftware Project Planning Size of five modules are Sersen it =4KLOC Command language inerpcer—=2 KLOC Fie input and output =1 Loc Cursor movement =2KLOC Screen movement aKLoc otal 1KLoc Software Project Planning Software Project Planning Let us assume that significant cost drivers are iL Required software reliably i high, be. 1.15, li Product complexity is high, Lo.1.15 ii, Analyst capability is high, 9.0.86 Iv, Programming language experience is low... ¥. Allether drivers are nominal EAP = 1.15x1.1540.261.07 2169 Software Project Planning (2) The initial fort estimate for the project is obtained from the following equation E =a (KLOOxEAF 3.2(12)'95x 1.2169 = 52.91 PM Developmenttime = C(E)! 5(62.91)02" = 11.20 (b) Using the folowing equations and refering Table 7, phase wise cost and schedule estimates can be calculated. E,= HE D,=1,D Since size is only 12 KLOC, itis an organic small model. Phase wise sffort stibuton is given below System Design = 0.18 52.91 = 8.485 PM Detailed Design 0.28 x52.91 = 13.756 PM Module Code & Test 0.425291 = 22.222 PM Integration & Test 0.165291 = 8465 Pm [Now Phase wise development time duration is. =0:19x11.29=2.145M 24x11.29= 2.709 M 9911.29 = 4.403 M 3ex11.29=2.032M System Design Detailed Design Module Code & Test Integration & TestSoftware Project Planning ‘cocomo- The following categories of applications / projets ae idemiid by COCOMO-II an are shown in fig. 4 shown Below: Software Project Planning ‘Application Composition Estimation Model Software Project Planning en [apace avon aesencomnn [ aie ey aie Software Project Planning iL Assess object counts: Estimate the number ot screens, reports and {8.6L component tha wil comprise this appeaton, i. Classification of complexity levels: We have 10 classy each ject instance into sgle, ecium and dficut complexity loves fepanding on values af ts charactors. 3-7 oy Mtn Dit 09 (a: For sneoneSoftware Project Planning Software Project Planning ———ae li. Assign complexity weight to each object: The weights are used ‘Nant’ |__edovre ofeien_____] for tives cbc pes Le, seen, repo and SGL components sing eSemne | Coane the Table 10 tart Single Sip Medi Objet Complexity Welght sed Sina ati Diet Tyre ere ee ae rn Moi iia ie a : a Table 9) Fo reson ad ai A A GL Component | — = 10 Software Project Planning iv. Determine object points: Add all tne weighted obec instances to ‘tone number and hs known aa ct point count \. Compute naw cbjoct points: We have o estimate the percentage reuse fo be scheved fn a project. Dapenang onthe perniage Fuse, tha new obec pots (NOP) are computed (object points *(100%reuse) NOP =n [NOP are the object poets that wil need tobe developed ander rom ‘he objet pont count because there may Be rouse Table 0: Compt wight or enc ve Software Project Planning i. Calculation of productivity rate: The productviy rate can be tulad a: Produsity ate (PROD) = NOP,Person moth ai aire ‘PROD worrP roan capa ies = var weSoftware Project Planning i.Compute the elf in Persons-Months: When PROD is known, \wo™may estimate ot n Person Months as: Nor efor in PM = Roo Software Project Planning Solution “This project comes under the category of application compesiion| estimation mad Number of screens = 4 with 4 views each Number of report ith 6 sections each From Table 9 we know that each screen will be of medium compleity and each reper willbe dificult complexity Using Table 10 of complexity weights, we may calculate object point count 24x242x8=28 24° (100-10) NOP = 100 =216 Software Project Planning Bvample: 49 (Consider a database aplication pojet wit the following carters: 1. The application has 4 serwens with 4 views each and data tbles for 3 servers and clients. 1, The aplication may generate two report of 6 sections each rom 07 data tables for two server and 3 chet, There i 10% reuse of object pias ‘The developer's expeience and capability inthe snl environment is low The maturity of eeganizaton in terms of capability is also low. Calculate the object pat cous, New object pots and effort to develop sucha projet Software Project Planning Table 11 gives the low valu of productivity (PROD) Le. 7. NOP Efforts in PM = anSoftware Project Planning The Early Design Model ‘The COCONO-I models ve the basa equation of he fxm PMs =A *(Siz0)® where Plonina = Etro th project in person months ‘A= Constant representing the nominal product, provisional Seto 25 B= Seale factor ‘Size = Software size Software Project Planning “The yale of can be auld 8-091 + 0.01 * (Sum of ating on sealing facies forthe projet)Software Project Planning Early design cost drivers ‘Thor ate sven ely design cos divers and ae given blow iL Prout Relish ant Complexity RCPX) 3 Roguived Reuse (RUSE) ii, Platform Dicuky (PDIF) in. Personnel Capability (PERS) Personnel Experience (PREX) i. Fails (FCIL) Vii. Sehodule(SCED) Software Project Planning Software Project Planning Post architecture coat avers “There te 17 com dives inthe Post Architecture model. These ae ated ona scale of 1106 a8 given below rs “The lis of seventeen cost vers given below 4. Reliabiiy Rogused (RELY) i, Databas Size (DATA) 4, Product Complexity (CPL) Iv, Requied Reusabilty (RUSE) Software Project Planning ‘Documentation (DOCU Mi Exceuton Time Constant (TIME) i Main Storage Constant (STOR) i Platform Volailiy (PVOL) |x, Analyst Capailiy (ACAP) Programmers Capability PCAP) i, Personnel Comtinity (PCON) i Analyst Expoience AEXP) i, Programmer Experience (PEXP) sv, Language & Tool Experience (LTEX) sv. Use of Software Tools (TOOL) vi, Site Locations & Communication Technology between Sites (SITE) sv Sobadule (SCED)Software Project Planning ‘Wapping of early design cost drivers and past architecture cost” drivers ‘The 17 Post Architecture Cost Drivers ae mapped 107 Eatly Design Cost Drivers and are given in Table 1 ie Je Mai Software Project Planning 1, Required Rouse (RUSE) : This ey design model cost diver is same as 14 Post achtece Couserprt. The RUSE rating levels ae (ah per Table 16) el =[=" | |i | ie vo Software Project Planning Product of cost divers for early design model 1. oduct Relisbiliy and Compleiy (RCPX): The eos iver combines fou Post Architecture cos vers which a RELY, DATA, CPLX and ocu Software Project Planning Platform Dlficulty (PDIF) : This cost diver combines TIME, STOR ‘nd PVOL. of Post Arcitectte Cost DrivesSoftware Project Planning iy, Personnel Capaiity (PERS) : This eost driver combines thee Post ‘Architecture Cost Daves, These drivers are ACAP, PCAP and PCON, Software Project Planning Personnel Experience (PREX) : This etl design diver combines thee ost Architecture Cost Drivers, which ate AEXP. PEXP and LTEX, Software Project Planning = = ee ‘ae [|= | [ae Cotecrarea | oe [ow | we | me | ee | | we ‘tec Ron Tene} 3] S|] oe me Software Project Planning Fates (FCI) This depends on two Post Architecture Cast Drivers, which are POOL atl SITE \iSehedule (SCED) : This ely design cos diver isthe same as Post ‘Architecture Counterpart and ating level are given below using table 16 Som Woe win | Mab | eoiieh aleSoftware Project Planning The seven cally design cost divers have been convened into numeric ‘lus with Nomina vale 1.0. These values are used forthe calultion ofa factor called “Effort multiple which the product of al Seve eal design cos drives. The muerte values te given in Table 15, coo TR TE Eames gnats wo Software Project Planning Software Project Planning “The carly design model adjusts the nominal effort using 7 effort mulls (EMS) Each effot multiplier (lo called dives) has 7 possible weights 26 fiven in Table 15, These factors are used forthe eatation of adjusted effort as given below: Pha PM pa Trt may very even up to 4008 fom PML Hence PMs ithe ine tune vale of effort inthe etl design phase Software Project Planning Example: 4.10 [A software project of application generator category with estimated 50, KLOG hast be developed. The scale factor (B) has low precedeniness, high development fexiiity and tow team cohesion ‘Other factors are nominal. The early design cost drivers lk platform itfeut (PDIF) and Personnel Capability (PERS) are high and others fae nominal, Calculate. the effort in person months for the dovelopment of the projec. Souton Hore B=091 + 001° (Sum of ating on scaling actors forthe projet) = 091 +001" (4.96 +205 + 4.24 + 4.38 + 4.58) $091 + 0.0120.29}1.1128 PMonna = Asi8)® = 25" (5)! = 194.41 Person mons “The 7ooat vers are OIF = high (1.29) PERS = nigh (0.89) RCPX = nmin! (1.0) [RUSE = nominal (1.0) PREX = nominal (1.0) CL. = nominal (10) ‘SCEO = nominal (1.0Software Project Planning Plaga PM a EEN, Software Project Planning Post Architecture Mode! ‘The post achtecte model isthe most died estimation model and is intented to be used when a sofware life cyele architecture has een ‘completed, This model is used inthe development and maintenance of sae prods i he agleion get sem ition = 194.41 *[1.20.x083) PM cys =PM yaya TEM, “ecator ” Y = 208155 Person montis EM: lft muti whihs the rods 7 cost dives “he 17 cot ives ofthe Pst Archictare model ae desi ia the ute Software Project Planning Software Project PlanningSoftware Project Planning Software Project Planning (or [ecm wo lE~ fe ie Software Project Planning Software Project Planning Product complexity is based on contol operations, computational operations, device dependent opeatios. data management operations sed tte mrice managenen peaos, Mole comple rung ae ven faz arc fone” fastttom [asia esann inner. sama [ages Janae [oss [ Calton of th rut of eff tot apr CM) Hence PM sce sect lf or en = Soca Naay | 3c)Software Project Planning Software Project Planning ire Project Plann oftware Prope a Software Project Planning RaineSoftware Project Planning ‘Schedule estimation Development ine can be caeulated using PML. 36 ey factor and the esd equation is naneni-senn 4 SCED% co 10M whore =constan provisionally st 103.67 TDEV gyn = Sse ein months with scheduled constrain aling factor PM jaa = tinted ort in Person months (after adjustment Software Project Planning Software Project Planning ‘Size measurement Size can be measured in any unit and the model can be calibrated ‘secordingly. However, COCOMO It details re: 1. Application composition model uses the size in objet pints i. The otter two models use size in KLOC ay design model uses unadjusted function pots. Those function poins sae converted into KLOC using Table 19, Post architecture model may Compute KLOC afer defining LOC counting rues I function points ate used, then use unadjusted fonction points ad convert it ato KLOC using Tae 19. Software Project Planning Tongues ‘SLOCUP, lade 7 [al Sha 2 [APL 22 lassen 320 [Assembiy (are) are PANSY Gui Turbo Baa et Basi-Conpied a Basi erred 8 fe 8 oa 2 Tongues ‘SLOCURP, ANSI Cobol aS 31 Fontan 77 105) Ft 64 oval 105 se 6a Tod a0 Pascal a Prog ee port Ganeraior a0 [Spresdsheet =Software Project Planning Example: 4.11 Cooder the software projet given in example 4.10, Size and sale fet (B) ae the sae, The Mentified 17 Cot divers are high reisbilty (RELY), Very high databuse size (DATA), high exceuton tine consaint (TIME). ‘ery high analyst capability (ACAP), high progsammers capability (PCP) “The oer cot divers are nominal, Calculate he effort in Peson-Months fo the development ofthe project, Software Project Planning Putnam Resource Allocation Model Norden of IBM ayleigh curve Model fora range of hardware development project. Persons —Time— he Rayleigh manpower loading curve Software Project Planning Solution Here Ba 1129 = 194.41 Person nth Pusan fT | = 1941 x (15 x L198 Lt 2067087) = 1941 xo.8ss = 17205 Person ots Software Project Planning Putnam observed that this curve was a close approximation at project level and sofiware subsystem level. No. of projects = 150Software Project Planning ‘The Norden / Rayleigh Curve “The curve is modeled by di real equation pate mt) = © = ohare u(t) dt te manpower uilizaon rate pet uni time parameter that fects the shape of he curve ea under curve in the itera (0, 1] lapsed time Software Project Planning ae" [1-2ar") “g'sme where maximum efor ate cuts Replace "for #in equation (2) Software Project Planning (On Integration om interval [0,1] yO =K Le] 2) Where yt): cumulative manpower used upt time t yiO)=0 yo)=k ‘The cumulative manpower is null atthe start of the project, and rows monotonically towards the total effort K (area under the Software Project Planning Replace “a with —* in the Norden/Rayleigh model. By ‘making this substitution in equation we have m0Software Project Planning me Person Time Gears) —> Influence of parameter ‘a onthe manpower Software Project Planning Example: 4.12 A software development projec is planned to cost 95 MY in a period fof | year and 9 months. Calculate the peak mani fof sftware team buildup. average rte Software Project Planning peak manning m() is obtained and denoted by m, k eal project cosveort in person-years iver time in yeas smi, = No.of persons employed tthe peak 71898 Software Project Planning Solution Software development cost k=95 MY Peak development ine 25 yeas Peak main 95 Toe 32.04 =33 persons Average ate of software team buildup m3 2. 33 s-srid K:py T: Years S=CK c ‘The trade off of time versus cost Software Project Planning Software Project Planning Devdopmen Subeyae Alt hat has been disused 5 far seated to projet fe cple as Project ie cycle ‘Spent by poet ce Project curve is the addition of two curves Manpower Development Test & Curve Validation CurveSoftware Project Planning +, my (t)=2kybte? yal)= Ky[he™] ‘An examination of mj) function shows a non-zero value of m, atime ty. “This is because the manpower volved in design & coding is still completing this activity aller € in form of rework due to the validation ofthe product. Nevertheless, for the model, a level of completion has to be assumed for development. It is assumed that 959% of the development will be completed by the time fy. Software Project Planning Relationship between Ki & K must be established, At the time of origin, both eyeles have the same slope. Software Project Planning We may say that > s 2h ‘T,g: time at which development curve exhibits a peak Software Project Planning ‘This does not apply to the manpower build up Dy K_K, p,=-K-Ke “ot Vor, Conte investigated that Larger projects > reasonable ‘Medium & small projects —> overestimateSoftware Project Planning Example: 4.15 [A software development requires 90 PY during the total development sub-ycle. The development time is planned fora duration of 3 yeas snd'5 months (Calculate the manpower cost expended until development time (b) Determine the development peak ime (6) Calculate the difficulty and manpower build up Software Project Planning ()Weknow fomeqution Vo ty = 1 = 3.41 /2.449 = 1.39 years “ve = 17 months Software Project Planning Solution (4) Duration = 1 yeas ‘We know from equation Ya(0) eo =095 Software Project Planning (©) Tol Manpower development Ky = ytg)10.95 =855/098=90 K=6K, =90x6=S40PY 540/(3.41)? =46 pesouvyeus 540/3.41)° =13.6 pesonsyeus?Software Project Planning Example:4.16 A software development for avionics has consumed 32 PY up to development cycle and produced a size of 48000 LOC. The development of project was completed in 25 ‘months. Calculate the development time, total manpower requirement, development peak time, difficulty, ‘manpower build up and technology factor. ‘Technology factor C=SK 3077 Software Project Planning Solution: Development time t, = 25 months = 2.08 years “G08 Software Project Planning Example 4.17 ‘What amount of software canbe delivered in | year 10 months in an ‘organization whore technology factor is 2400 if a total of 25 PY is Permitted for development effortSoftware Project Planning Software Project Planning Solution: Example 418 ty =1.8years ‘The software development organization developing real time . software has been assessed at technology faetor of 2200. The Ke s25Py maximum value of manpover build up for this type of K =25x6=150PY software is D,=7.5. The estimated size to be developed is © = 2400 s=55000 Loc Weknow $= CK "1,4 (@)Determine the total development ago 513 «2.18 27020 Loc development manpower cost, development peak m time, the total the difficulty and the (b) The development time deter sd in (a) is considered 100 long. Its recommended that it be reduced by two months. ‘What would happen? Software Project Planning Software Project Planning Solution since © Wehave $=CK'%r,! 3 ty=3 years GE K =Dy) =7.5x27=202 PY 202 hich is also equivalent to Total development manpower cost K,=22 06 D = Doty = 22.5 persons / year 3 y= tha 4 = 12 years “6 V6Software Project Planning M(t) = 2ky bie” Ya) = ky(-e"") Here od Peak manning =m, =Dt,ye'”? =225x 1.2.x.606 = 16 persons Software Project Planning Software Project Planning IIL If development time is reduced by 2 months Developing Producing siw at higher less software ‘manpower build-up Software Project Planning (@ Increase Manpower Build-up ate} Now ty=3 years 2 months =2.8 years p, =03)'103)" 1.6 persons years k= Dyth =254PY 254 _aapy 6 D= Dj, = 325 persons / year ‘The peak time is t,y= 1.14 years Peak manning Note the huge increase in peak manning & manpower cost,Software Project Planning i Produce Less Software Productivity versus difficult vample 419 ‘A stand alone projeet for which the size is estimated at 12500 LOC is to be developed in an environment such that the technology factor is 1200. Choosing a manpower build up Calculate the mini total development man power cost, the difficulty, the peak manning, the development peak time, and the development productivity. Software Project Planning S| 5x(2.8)' =10119.696 le [2] -2u00099 ie} ‘Then for (C=2200 $=47586 LOC Software Project Planning Station Sie) = 125001L0C Technology factor) = 1200 Manpowerbuildyp(D,)=15 Now S=CK ry? Sponge a Ki Also we know D, = ‘Substituting the values, we get [ese T 1, $185 yearsSoftware Project Planning (Gi) Total develop Hence K=15i) =15(1.85)'294.97 PY i) Dittcutty Software Project Planning (vi) Development Produetiv No.of lines of code (8) - ‘effort (K.) Software Project Planning (iv) Peak Manin, Software Project Planning Software Risk Management ‘> We Software developers ate extremely optimists ‘> We assume, everything will go exactly as planned. > Other view ‘not possible to predict whats going to happen ? Software surprises ‘Never good newsSoftware Project Planning Risk management is required to reduce this surprise factor Dealing with concern before it becomes a crisis. ‘Quantify probability of failure & consequences of failure, Software Project Planning Software Project Planning What is risk ? ‘Tomorrow's problems are today’s risks. “Risk is a problem that may cause some loss or threaten the success of the project, but which has not happened yet”. Software Project Planning Risk management is the process of identifying addressing and eliminating these problems before they can damage the project, ‘Current problems & L Potential Problems ‘Typical Software Risk Capers Jones has identified the top five risk factors that threaten projects in different applications 1. Dependencies on outside agencies or factors. + Availability of trained, experienced persons + Inter group dependencies + Customer-Furnished items or information ‘+ Internal & external subcontractor relationshipsSoftware Project Planning Software Project Planning 2. Requirement issues + Lack of lear product vision Uncertain requirements + Lack of agreement on produet requirements + Unprioritized requirements ‘Wrong product ‘+ New market with uncertain needs or + Rapidly changing requirements Right product badly ipa emangng ee + Inadequate Impact analysis of requirements changes Either situation results in unpleasant surprises and saguate Impact analyss of reat ss unhappy customers, Software Project Planning Software Project Planning 3. Management Issues 4. Lack of knowledge Project managers usually write the risk management + Inadequate raining plans, and most people do not wish to air their ‘weaknesses in public + Poor unferanding of methods, to, and + Inadequate planning fechniques + Inadequate visibility into actual project status + Inadequate application domain experience + Unclear project ownership and decision making + New Technologies + Staff personality conflicts + Ineffective, poorly documented or neglected + Unvealistic expectation processes + Poor communicationSoftware Project Planning 5. Other risk categories + Unavailailty of adequate testing facilities + Turnover of essential personnel + Unachievable performance requirements ‘+ Technical approaches that may not work Software Project Planning Risk Assessment dentification of risks Risk analysis involves examining how project outcomes change with modification of risk input variables. Risk prioritization focus for severe risks, Risk exposure: It is the product ofthe probability of incurring a loss due tothe risk and the potential magnitude of that loss. Software Project Planning Risk Management Activities Risk [dentification hae Risk Analysis Apesment Risk Prioritization Risk [Management TRisk Management Planning Risk Monitoring Fig. 9: Risk Management acts Risk Resolution Software Project Planning Another way of handling risk isthe risk avoidance. Do not do the risky things! We may avoid risks by not undertaking certain projects, or by relying on proven rather than cutting edge technologies.Software Project Planning Risk Control Risk Management Planning produces a plan for dealing with ‘each significant risks, > Record decision in the plan. Risk resolution isthe execution of the plans of dealing with ‘each risk. Multiple Choice Questions {slit anton poi afm group {ctrl fnton pit wer oop {46 Fanon pit canbe acted by tur "car (iy urrs Fac (e} UEP Coat (@) UEP Podatviy (a) Functor pointe (6) Nene fayigh care (6) Pama ery of stare unagement Multiple Choice Question ‘Nove: Choose mot appropriate answer of ie Tolowing questions 464 Faction pit analy (FPA) tod decompones he syste it fain Multiple Choice Questions ()BBeuer (Rae Gp (ey EtaaKLOC Pat (E2AKLOC) PM (eyTine (Gh Nowe of te above In fncton pit analysis aber of Comply austen factorateMultiple Choice Questions (shop gprench {by Aesth pouch (6) tsp prc (a) Top down ape {a} Orsini (Semis (eh mbes (a) None othe bone a) Unset of Maryland (0 Uninet of Southern Cairn ro (Are Tell abe “418 Which ne not Category of COCOMO.L (a) End User Progra (blurt Secor (eh Regimen Sector (a) Sytem neratoe Multiple Choice Questions fo case C=sKe! fo casey we Multiple Choice Questions 4.22 Panam source lation mel Rayleigh curves modded by te equation 1b) mey= 2k G4) my =2KH6 ‘Exercises 1 Vara in ding twa oR pling? 42 Describe any oo software size estimation tchniqus. 443 A proposal is made to count the size of programs by oumber of Semicaons, excope those oscuting with tral stings. Discs the Stengihe and weknawss to hs sé messue when compared with fe ines of ead count 44 Design a LOC counte for counting LOC automatically. Ist tnguage Apoidea! What ate te inition of such acounrs 45 Compute the fonction point value for a project with the flowing formation domain charters, Number of wer inputs = 30 [Number of wer outputs = 42 [Number of extemal interfaces = Assume tat all complexity adjustment values are moderate‘Exercises 46 Explain the concept of function points, Why FPs are becoming accion indus? 447 What ape the size meee? How i function point mesic advantageous (ver LO mec? Expats m " 448 Is it posible to enimate sofware size before coding? Justify your answer swath euable example 4.9 Describe the Avec’ function count method wit sutble example. 4.10 Compute he funeion pine FP fora parol program hat reads 2 ile of ‘ples and of information arth’ cument swan pans hdgue forall the enployeen The progeam fs eapabe of andy an aefcive command’ 10 pant af iavidily” guested cheque timed m= boss “ € “xercises ‘Exercises 4411 Assume tat the previous payroll program is expected to cead a le contain information abou all the Ghaguts aa hive bee piste The fle suppoud tobe pind and alo wed bythe program aba ine fi. fo ie tha pars pal ekpenn of he cent {hs programy haufy the iiference between te function pou of is Plant an previus one by conskering how the complet of the frogram it aftcted by. adding the reguvement of inrfaing with nates appization in thisease ise, 4.12 Esplin the Walson & Felix model snd compare with the SEL made 4.13 The sie ofa software product to be developed as been esate 10 be "2200010 Prodi the manpower cos (lft) by Walston Fels Medel id SEL del 4.14 database system io be developed. The effort hasbeen estimated wo ‘be 100 Pers Monts. Calculate the numer of nes of cade and productivity in LOCIPeson- Mout ‘Exercises 4.15 Discuss various pes of COCOMO mode, Explain the phase wise UUntsiution of tia isin he pes 4.16 Explain ll he levels of COCOMO mode, Assume tat the size of an logge stare euch Daca exited to be 33.000 lines of code ‘Beermine the efor equted to developed the sofware product and the Domina development tine " 4.17 Using the hase COCOMO model, under all de. operating mode, ‘striae the performance flaton Tor the rao of delivered source cae lines per posen-meonth of eff. Determine the reasonableness of tis rslaton fr several eyes of software projets 418 The effor disuibuion foe a, 240 KLOC, eganie, mode software evelopment project is product design 124, dated denn 24%, code ‘bu te nate at fs 28, How woud th ll ‘knges, fom ow wo highs fect he phase dstibulion of effort an th toaleffr analyst capaiiy use of moder programming languages ‘oaied reiabdiy. soiree voisiliy? "Panne lana 4.19 Specify, design and develop program that implements COCOMO. ‘sig san ude, exon program that can eed 8 Plannin oo 4.20 Suppose sysem for office automaton isto be designs Ii clear om feguigeents a here wl be five modules of si 03 KLOC. 13 KLO¢, 30 LOG: 0 RLOC and 2.0 KLOC respectively. Comply 2nd sclbilty rojitemeats are hgh. Bograniner’s capa and perience is low All eter factors are of toma ing. Use COCOMO. ‘obisItodeterne overs cow and schedule ext, Alo Caco thes a schedule esses for different phases. 4.21 Suppose that projet was estimated to be 600 KLOC. Caleulate the ilo? and development me foe each ofthe tnce modes. organic, ‘Sntdctached snd embedded the COCOMO-Itin detail. What types of categories of oj me‘Exercises {Bi Dac ts Intasmactare Sst of COCOMOT, “4.24 Desrite varios stages of COCOMO-H. Which stage is mote popular ‘nd why? 425 sofware projet of application generator category with eximated size ‘of 100 KLGC has tobe developed. The see Taco (B) bas ish coders, high development flexibly. Othe factors ae nomial BcScout der rgb sca aut dab isc gh Personel capably high-anlyt copay. The oder cost divees ae amin Cabal the Stat it Par Moa ore Svelopnar of the poet, 4.26 Explain he Panam reource allocation mods. What ate the liniations ‘tts 4.27 Describe the tae-off between time versus cost in Putnam resource ‘cation ae 4.28 Discuss the Putnam sesouces allocation model. Detve the ie and effort equations. € “xercises 434 Nat is ak expoue? What ects an be ws to cont ech 4.38 What isk? Is i economical todo risk management? What isthe effet ‘ofthis stvty onthe overall cata he projet? 4.36 -Thee ae significant risks even in student projets. Analyze a student projet andi all he sk pe ™ ‘Exercises 429 Assuming the Pumam model with S=100.000 ‘Compute evelopment ume’, and manpower developinent Ky 430 Ohsin software productivity data fortwo or te software development [poarams. Use several cost estimating Medes discussed i this cpt ow tothe ests compare wih actual poe ests? 431.1 sem odd that cost and size estinstes ae developed dung software jet pinning ore deed sare oquenont anager dough Ris"heen conducts Why! do. we think this ts done? Ate thee ‘Srumsunces wher shou ae be done! 432 Discus typical software sss, How staff ummover problem affects sottware projects? 4.38 What ae ss management ctivs? eit possible to pire the isk?