APPLICATION OF NET PRESENT EARNED VALUE METHODOLOGY IN CONSTRUCTION PROJECTS

Transcription

1 Inernaional cienific Conference People, Buildings and Environmen 2014 (PBE2014) Ocober, 2014, Kroměříž, Czech Republic, KEYNOTE LECTURE APPLICATION OF NET PREENT EARNED VALUE METHODOLOGY IN CONTRUCTION PROJECT Aleksander rdić 1 *, Jana Šelih 1 1 Faculy of Civil and Geodeic Engineering, Jamova cesa 2, 1000 Ljubljana, lovenija Absrac Uncerainies are an ineviable par of a consrucion projec and may lead o ime and/or cos deviaions from he original projec plan. Boh are highly undesirable from he clien's as well as from conracor's side. Therefore, during projec execuion, projec manager wishes o conrol he risk of no achieving he iniial plan. Earned Value Analysis (EVA) is one of he mos frequenly used mehods o carry ou his ask. EVA, however, does no accoun for he ime value of money which is imporan for long-erm projecs of high value, such as consrucion projecs. To overcome his drawback of EVA, research presened in he paper proposes an efficien approach by which EVA and Ne Presen Value (N) are combined ino a single mehodology. The proposed mehodology uses he Earned chedule (E) for he esimaion of projec duraion. The use of he new mehodology is demonsraed on a seleced case sudy from consrucion field. The resuls show ha a relaively simple and efficien ool ha can be successfully applied for long erm, cos inensive projecs, has been developed. Key words Earned schedule; earned value analysis; ne presen value; projec conrol To cie his paper: rdić, A., Šelih, J. (2014). Developmen of Ne Presen Earned Value Mehodology and is Applicaion in Consrucion Projecs, In conference proceedings of People, Buildings and Environmen 2014, an inernaional scienific conference, Kroměříž, Czech Republic, pp , IN: *Corresponding auhor: Tel.: ; address: aleksander.srdic@fgg.uni-lj.si 16

2 Inernaional cienific Conference People, Buildings and Environmen 2014 (PBE2014) Ocober, 2014, Kroměříž, Czech Republic, 1 INTRODUCTION Projec execuion and is monioring and conrol are he longes phases in projec life cycle. In his projec phase, he occurrence of unforeseen circumsances and changes in he projecs scope can ofen be noiced. This is paricularly valid for consrucions projecs. As a consequence, monioring of projec developmen sars on consrucion sie, by gahering coss and acual realisaion dae of aciviies. Any deviaion from he conrac-baseline schedule and budgeed coss should be idenified and hen analysed in order o be able o answer wo quesions, ha are, from clien s poin of view, he mos relevan: When, and a wha coss he projec will be finished. The answers are based on he projec saus a he conrol dae, which is a combinaion of finished projec aciviies and associaed acual coss. Analyses of acual daa provide also he baseline for he esimaes of fuure projec realisaion. The mos esablished mehodology for projec monioring and making decisions on he op projec managemen level is Earned Value Analysis (EVA). This mehod is measuring projec performance and progress from he viewpoin of achieving he scope, schedule and coss, and is inegraing hese parameers ino a single sysem. In order o ensure successful projec execuion, special aenion should be placed o he projec feasibiliy phase, where goals of he projec are already clearly defined, and risks are evaluaed. Two main decision crieria employed in his phase are projec Ne Presen Value, N and Inernal Rae of Reurn, IRR. Despie significance of hese wo indicaors, boh of hem are mainly ignored during consrucion phase. I is clear ha EVA-based key indicaors are no sufficien for efficien projec monioring. The mehodology proposed in his paper ries o overcome his defici. Coss performance index (CPI) convenionally used in EVA for he esimaion of he remaining cos, and esimaion of he projec end, based on schedule performance index () ha is calculaed wih Earned chedule (E) mehodology are used. To evaluae he esimaed projec financial oucome, all realised and esimaed remaining coss are ime phased o corresponding inervals and discouned o he presen value, and projec esimae of he ne presen value can be deermined. In addiion o he esimae of he profi ha is a convenional EVA indicaor, projec revenues have o be included as an indicaor as well. The purpose of he research presened in his paper is o idenify challenges associaed wih he use of Earned Value Analysis (EVA) in consrucion projecs, and o esablish a model by which Ne Presen Value (N) is accouned for wihin Earned Value Analysis. 2 TATE OF THE ART Earned Value Analysis is well acceped mehodology ever since 1979 when U.. Deparmen of Energy proposed he Earned Value ysem as complemenary conrol aid o he classical scheduling mehods - i.e., o he Criical Pah Mehod (CPM) and he Program Evaluaion and Review Technique (PERT). Aside from he problems associaed wih poor cos or duraion esimaes, boh mehods, CPM and PERT, have srong limiaions when rying o describe he variable naure of projecs, especially consrucion projecs [1]. From he inroducion of EVA onward, researchers are predominanly focused on enhanced reliabiliy of projec performance indexes [2], [3], [4]. As a consequence, addiional parameers, such as Earned chedule [5], [6], relaionships beween facor [7] or implicaion of differen echniques for predicing [8], such as regression [9], Neural Neworks [10] and Kalman filer [11] were developed. 17

3 Inernaional cienific Conference People, Buildings and Environmen 2014 (PBE2014) Ocober, 2014, Kroměříž, Czech Republic, Anoher aspec of research ha is currenly being underaken in his area is incorporaing projec-relaed uncerainy ino EVA as a par of he projec risk managemen. Research endeavours ha address his issue are predominanly based on fuzzy approach [12]. Recenly, addiional projec success indicaors, e.g. he esimaed projec N during projec execuion have been proposed [13], o be considered wihin EVA. 3 METHODOLOGY 3.1 Fundamenals of Earned Value Analysis EVA is used o measure he performance of a projec. I provides a sysemaic approach o inegraion and measuremen of scope, cos and schedule accomplishmens in a projec. The mos imporan indicaors employed in he analysis are CPI (coss performance index) and (schedule performance index). The efficien use of EVA requires ha projec WB (Work Breakdown rucure), CB (Cos Breakdown rucure) and feasible projec schedule wih reliable coss esimae are carefully prepared. Gahering of aciviies' saus and relaed incurred cos needs o be accurae, herefore significan effor is required o carry ou his ask. However, afer his sage is compleed, he analysis is simple and quick. Key indicaors Projec conrol mehodology is based on analysis of ime phased projec coss (fig.1). The saring poin for he analysis and calculaion of projec performance indicaors is projec baseline cash flow BCW() (Budgeed Cos of Work cheduled), ha represens cumulaive budgeed cos a given ime during projec execuion. Fig. 1: Basic elemens of EVA EVA provides wo key indicaors: CV() and V() - cos and schedule variance a ime of conrol, = conrol. Boh indicaors describe he calculaed difference owards BCW, herefore hey are defined as: CV() ACWP ( ) BCWP ( ) (1) 18

4 where: Inernaional cienific Conference People, Buildings and Environmen 2014 (PBE2014) Ocober, 2014, Kroměříž, Czech Republic, ACWP () Acual Cos of Work Performed, and BCWP() Budgeed Cos of Work Performed V() BCWP ( ) BCW ( ) (2) Projec performance assessmen, esimaion of ime of projec compleion and cos a compleion are assessed by he following performance indexes a a given ime : - Cos performance index, CPI BCWP() CPI() (3) ACWP() When he cos of compleing he work is righ on plan, CPI equals o 1. CPI < 1 indicaes ha cos of compleing he work complies o he planned value, and CPI >1 indicaes ha he cos is smaller han planned. - chedule performance index, BCWP() () (4) BCW() When he projec is on schedule, equals o 1. < 1 indicaes ha he projec is behind he schedule, and >1 indicaes ha he projec is ahead of he schedule. 3.2 Earned chedule Expression of chedule Variance (eq.2), where ime performance is based on cos raio, seems o be somehow srange, because ime and cos aren same caegories. Many auhors ([2], [3], [4], [5], [6]) recommend use of Earned chedule approach o deermine V and, which are based on ime of conrol ( conrol ) and ime of when BCW=BCWP ( ) (fig. 1). where: V V ( ) conrol conrol BCW BCW (5) BCW ( ) (6) conrol conrol schedule variance based on Earned chedule mehod schedule performance index based on Earned chedule mehod 3.3 Ne Presen Value (N) In projec managemen, he ime and money are differen caegories inerconneced by he ineres rae. Ne presen value (N) is he sum of ime phased coss and revenues discouned o he presen value, over he ime period under consideraion (eq. 7): where: ime period couned from ime of presen value C cos or revenue in period r discoun rae for he considered ime inerval T P C N (7) 0 (1 r) 19

5 Inernaional cienific Conference People, Buildings and Environmen 2014 (PBE2014) Ocober, 2014, Kroměříž, Czech Republic, Merging N ino EVA Basic EVA does no ake ino accoun he ime value of money. Consequenly, in long erm consrucion projecs, where delays wih duraion of several monhs can occur, he mehodology does no exhibi proper cos performance index. Even if acual cos doesn overcome he budgeed value, i does no mean ha he projec will be finished wih planned profi when ime value of money is aken ino he accoun. In addiion, EVA has a sysemaic failure: a he end of projec, earned coss are he same as budgeed. This means ha despie he evenual delay, a he end is always 1, and does no show he discrepancies from he original plan ha occur during execuion. The mehod proposed in his work ries o overcome he above presened lack of crucial informaion by comparing presen acual and earned value wih budgeed presen value. Early and lae projec schedule Before ime value of cos and revenue is aken ino accoun, differen possible iming of aciviies realisaion should be clarified. Usually, projec baseline schedule is based on early daes of aciviies realisaion. The main reason for his decision sems from he desire o schedule aciviies wih maximum possible floa ime, so ha evenual delays ha may occur during projec execuion can be compensaed. Insigh ino available floa can be indirecly presened wih early BCW E and lae BCW L cumulaive diagram, presened in Fig. 2a. Fig. 2: Cumulaive budgeed cos a projec aciviies early and lae realisaion If we ake ino he accoun ha delayed occurrence of he seleced cos means greaer value L discoun, value of cos for aciviies realisaion a lae ime - BCW are lower han E of early ime - BCW (fig. 2b). A conflic siuaion regarding he aciviies and projec ime reserve (floa) appears. Real world projec managemen is compromise, so of budgeed cos work performed, BCWP, should be compared wih boh - lae and early budgeed presen value (eq. 8, 9). E () BCWP() (8) E BCW() 20

6 Inernaional cienific Conference People, Buildings and Environmen 2014 (PBE2014) Ocober, 2014, Kroměříž, Czech Republic, L () BCWP() (9) L BCW() The meaning of he performance indexes defined by eqs.(8) and (9) is presened in Table 1. Tab. 1: projec ime performance Indexes Value E > 1 E < 1 and L > 1 L < 1 Meaning ahead of schedule sill on schedule, floa decreased behind schedule Cos performance index Cos performance index, CPI, can be calculaed in he same manner as in exising EVA (eq. 3). However, even if CPI < 1, i is possible ha CPI > 1. If, a he same ime, he relaions E < 1, and L > 1 apply for hese wo performance indexes, i means ha he projec will be finished wih expeced N. Esimae a compleion During projec execuion, here are anoher wo very imporan quesions for he projec manager: when will he projec be finished and a wha coss and profi? imple comparison of coss and revenues does no show he real projec success. If revenues occur a he end of projec (e.g. build and sell aparmens on he marke), only he projec s N is relevan for is success evaluaion. In such case, despie he fac ha lae accruing of coss decreases heir presen value, he presen value of revenues is also grealy decreased due o he projec delay. Esimae of projec duraion is herefore a crucial parameer in he calculaion of N, by which projec profi can be esimaed. Wihin EVA, projec duraion esimae is based on and original (baseline) duraion. The following equaion applies: where: ED OD ED (10) and OD are esimaed, and original duraion, respecively. This esimae is ofen inappropriae for he consrucion projec, due o he fac ha major delays frequenly appear during he firs hird of he projec. Typical aciviies where delays can occur in his par of he projec are, for example, esablishing conracors coordinaion, or dealing wih unforeseen geoechnical condiions. These delays canno be spread linearly over he whole projec duraion, herefore is no a suiable indicaor for assessing he consrucion projec performance. also does no exhibi realisic performance during he las hird of projec execuion, due o he sysemaic failure of EVA already menioned. Therefore, he presen research proposes an improved mehodology ha is based on Earned chedule mehodology. The esimaes of projec finish and he remaining duraion of he projec are formulaed as: - simply adding curren delay esimaed duraion, ED + V o he original projec duraion ED T V P T planned, o obain he P planned (11) 21

7 Inernaional cienific Conference People, Buildings and Environmen 2014 (PBE2014) Ocober, 2014, Kroměříž, Czech Republic, - deerminaion of projec duraion based on Earned chedule mehod s : o o L, - according o lae daes (BCW L ) P T xl planned ED (12) L,,E - according o early daes (BCW E ) P T planned ED (13),E Esimae of remaining coss - ETC (Esimae o Complee) flow When ime value of fuure coss is considered, i is necessary o deermine he cos flow unil he end of projec. Esimaion of he remaining cos (ETC) is calculaed as he difference beween oal budgeed coss and budgeed cos of work performed, correced wih curren cos performance index ( BAC BCWP) ETC CPI The remaining coss are spread on ime inerval beween ime of conrol and esimaed duraion, as schemaically presened in Fig. 3. (12) Fig. 3: Cumulaive curve of acual and esimaed remaining coss Presen Value of combined ACWP and ETC Based on combinaion of acual and esimaed o compleion coss flow (Fig.3), heir presen value can be deermined by he Eq.(7) and use of differen esimaed duraion ( ED xl, ED, ED ). Presen values,, of 3 differen scenarios are labelled as EAC, xl EAC and EAC. 22

8 Inernaional cienific Conference People, Buildings and Environmen 2014 (PBE2014) Ocober, 2014, Kroměříž, Czech Republic, Esimaed Projec N a compleion By adding ime phased revenues based o projec esimaed compleion, he proposed mehodology provides 5 differen projec ne presen values: - BE P N projec early baseline N - BE P N projec lae baseline N P N P N xl P N projec projec projec EAC EAC xl EAC R R xl R 4 CAE TUDY The pracical use of proposed mehodology is presened for he following case sudy -residenial building projec. 4.1 Baseline daa (Fig. 4) Coss and revenues are presened as of he oal coss. The seleced annual discoun rae is 8%. The following daa were provided: OD= 30 monh BAC= 1.000,0 uni R=1.200 uni a he end of projec; Profi= 200 uni 4.2 Expeced N of projec Ne presen values were deermined for he opions Early schedule and Lae schedule. The following values are obained: E - Early schedule : BAC 903,7 R E 983,6 P 983,6 903,7 79,9 L - Lae schedule : BAC 897,3 R L 983,6 P 983,6 897,3 86,3 4.3 Projec schedule performance conrol Projec performance is evaluaed from he viewpoin of cumulaive coss developmen wih ime. The conrol poin is se a Tconrol=15 h monh. Fig. 5 presens he coss flow from projec sar o conrol ime. Comparison of BCWP wih BCW E shows ha i should be earned a 13,6 monh from he sar, which means ha he value of V is -1,4 monh, according o he Earned chedule mehod. V based on comparing wih BCW L is less criical, only 0,3 monhs. N N 23

9 Inernaional cienific Conference People, Buildings and Environmen 2014 (PBE2014) Ocober, 2014, Kroměříž, Czech Republic, coss [ ] revenues [ ] 1300,0 1200,0 1200,0 1100,0 1000,0 1000,0 900,0 800,0 800,0 700,0 600,0 600,0 500,0 400,0 400,0 300,0 200,0 200,0 0, R 0,0 0,0 0,0 0,0 0,0 0,0 0,0 0,0 0,0 0,0 0,0 0,0 0,0 0,0 0,0 0,0 0,0 0,0 0,0 0,0 0,0 0,0 0,0 0,0 0,0 0,0 0,0 0,0 0,0 0,0 1200, BCW E 5,0 15,0 25,0 40,0 55,0 80,0 105,0 135,0 165,0 205,0 245,0 295,0 345,0 405,0 465,0 535,0 605,0 655,0 705,0 755,0 795,0 835,0 875,0 905,0 930,0 950,0 970,0 985,0 995,0 1000, 1000, BCW L 3,0 11,0 19,0 31,0 44,0 64,0 86,0 111,0 136,0 166,0 201,0 246,0 291,0 341,0 396,0 458,0 523,0 573,0 628,0 683,0 733,0 785,0 830,0 868,0 901,0 928,0 955,0 973,0 989,0 1000, 1000, Fig. 4: Cumulaive coss a early/lae ime and revenue a he end of projec 100,0 0,0 ime coss [ ] 600,0 500,0 400,0 300,0 200,0 BCW E BCW L ACWP BCWP 465,0 420,2 396,0 385,0 100,0 0, ime Fig. 5: Cumulaive coss a conrol ime Projec performance is assessed by several indexes, as presened in Table 2 and figure 5. Resuls of he analysis show ha a he ime of he assessmen, he acual coss are exceeding he planned ones, bu esimaed coss are lower han revenue and profi is sill expeced. Furher, all obained values for he indicaors, ED, V, and ED indicae ha he projec is behind he schedule. According o EVA, expeced delay is abou 6 monhs while Earned chedule shows ha he esimaed delay is only 3 monhs when compared o early schedule and almos no delay (0,6 monh) when compared o he lae schedule. Considering he fac ha he conrol ime is in halfway of projec duraion and approximae half of floa is already consumed, he esimaed projec delay is abou 4,5 monhs. 24

10 Inernaional cienific Conference People, Buildings and Environmen 2014 (PBE2014) Ocober, 2014, Kroměříž, Czech Republic, In order o cope efficienly wih uncerainy, several differen scenarios (differen EAC and ED) need o be produced. Only hen, a beer insigh ino possible N (Fig. 6) can be gained. Tab. 2: projec performance indicaors Indicaors Value Meaning CPI EAC ED V ED E V L V E L ED ED ED P N P N xl P N Earned Value Analysis 0,916 Coss overrun 0,828 Behind schedule 1.091, ,4=108,6 (Profi decrease) 36 mo Esimaed projec delay 6 mo. Earned chedule 1,4 mo Curren delay 1,4 mo 0,91 Behind schedule 33,09 mo Esimaed projec delay 3 mo. Proposed mehodology (EVA+E+N) 1,4 mo Curren delay 1,4 mo behind early schedule 0,3 mo Curren delay 0,3 mo behind lae schedule 0,91 based on E early - delay 0,98 based on E early-small delay 30,85 mo Esimaed projec delay 1 mo. 33,09 mo Esimaed projec delay 3 mo. 30,61 mo Esimaed projec delay 0,5 mo. -2,3-13,1-1,1 Negaive projec N no profi Negaive projec N loss Negaive projec N no profi 5 DICUION The resuls obained by using he combined Earned Value Mehod and Ne Presen Value concep show ha in large long lasing projecs, he use of his mehod is jusified. Cumulaive coss presened in Fig.4 show ha significan differences may occur during projec execuion. This may no necessarily mean ha he projec is a risk of exceeding he deadline, or he conracual cos. Neverheless, i can affec he cash flow a a given momen. As a consequence, i is clear ha EVA mus be supplemened by N, especially when projec are cos- and imeconsuming. 25

11 Inernaional cienific Conference People, Buildings and Environmen 2014 (PBE2014) Ocober, 2014, Kroměříž, Czech Republic, coss [ ] 1200,0 1000,0 BCW E BCW L ACWP 1100,0 30,61 ; 1091,4 33, ,4 800,0 600,0 BCWP EAC[ETC/] [ED xl] EAC [ETC/CPI][ED ] EAC [ETC +CV][ED ] 1050,0 1000, ,0 33,09 ; 1035,2 400, , 385,0 950, ,0 0,0 ime [mo] Fig. 6: Cumulaive expeced coss, based on projec performance a conrol ime Tab. 3: Projec N wih respec o differen discoun raes and projec performance indexes N P N P N xl P N Tp Discoun rae [monh] [% ] 8% 7% 6% +0,85 2,83-2,3 9,5 21,8 +3,09 10,29-13,1-0,3 13,2 0,61 2,04-1,1 10,6 22,7 Fig.5, where cumulaive resuls a conrol ime are presened for differen alernaives, shows ha discrepancies can be crucial for he analysis of pas resuls, as well as for he predicion of fuure rends. Values for projec performance indicaors colleced in Table 3 show ha for he ime considered, he projec is behind schedule and ha i exhibis a cos overrun. This observaion should no, as already discussed, be aken as he final oucome; he projec needs o be compleed in order o make he final evaluaion. Finally, i should be noed ha he calculaed N depends upon he assumed discoun rae, and upon he ime-o-compleion index value. 6 CONCLUION Large, cos and ime consuming consrucion projecs require specific approaches in all projec phases. Ofen, significan aenion is placed o he planning sage, bu execuion and conrol sage are negleced due o ime pressures ha ofen arise in he consrucion sage. The proposed Ne Presen Earned Value Mehodology can faciliae conrol ha needs o be applied during execuion. When i is used, special aenion should be paid o he ime inervals for which i has been shown ha he suiabiliy of he proposed mehod is low, i.e. firs and las 26

12 Inernaional cienific Conference People, Buildings and Environmen 2014 (PBE2014) Ocober, 2014, Kroměříž, Czech Republic, hird of he oal projec ime. However, when hese specific opics are adequaely addressed, he mehodology can be exremely useful for overall projec conrol. REFERENCE [1] Barraza G., Back W., Maa F. (2000). Probabilisic monioring of projec performance using curves. J. Consr. Eng., April, pp [2] Lipke W. (2012). peculaions on Projec Duraion Forecasing. PM World Today, 14(3). [3] Czarnigowska A., Jaskowski P., Biruk. (2011). Projec performance reporing and predicion: exensions of earned value managemen. In. J. Bus. Manag. ud., 3(1), pp [4] Lipke W. (2014). Examining Projec Duraion Forecasing Reliabiliy. PM World J., 3(3), pp. 1 10, [5] Görög M. (2009). A comprehensive model for planning and conrolling conracor cash-flow. In. J. Proj. Manag., 27(5)5, pp [6] Vandevoorde., Vanhoucke M. (2006). A comparison of differen projec duraion forecasing mehods using earned value merics. In. J. Proj. Manag., 24(4), pp [7] Cheng M. Y., Huang C. C.,Van Roy A. F. (2013). Predicing Projec uccess in Consrucion using an Evoluionary Gaussian Process Inference Model. J. Civ. Eng. Manag., 19(1), pp [8] Roodposhi F. R., Pakmaram A., Alamdari B. K. (2012). Forecasing Final chedule and Cos of Projecs by Using aisical and Mahemaical Mehod. In. J. Appl. Oper. Res., 1(3), pp [9] Narbaev T., De Marco A. (2014). An Earned chedule-based regression model o improve cos esimae a compleion. In. J. Proj. Manag. [10] Iranmanesh., Zarezadeh M. (2008). Applicaion of Arificial Neural Nework o Forecas Acual Cos of a Projec o Improve Earned Value Managemen ysem. Proceeding World Acad. ci. Augus, 32, pp [11] Kim B. C., Kim H. (2014). ensiiviy of Earned Value chedule Forecasing o -Curve Paerns. J. Consr. Eng. Manag., p [12] Moslemi Naeni L., hadrokh., alehipoura. (2014). A fuzzy approach for he earned value managemen. In. J. Proj. Manag., 32(4), pp [13] Jordanger B. I., Klakegg O. J. (2013). Value Managemen Beyond Earned Value. PM World J., 2(2), pp