Journal of Cleaner Production 39 (2013) 329e337 Contents lists available at SciVerse ScienceDirect Journal of Cleaner Production journal homepage: www.elsevier.com/locate/jclepro Economic viability analysis of a construction and demolition waste recycling plant in Portugal e part II: economic sensitivity analysis André Coelho, Jorge de Brito* Department of Civil Engineering, Architecture and Georesources, Instituto Superior Técnico, Technical University of Lisbon, Av. Rovisco Pais, 1049-001 Lisbon, Portugal a r t i c l e i n f o a b s t r a c t Article history: Received 16 September 2011 Received in revised form 31 March 2012 Accepted 3 May 2012 Available online 21 May 2012 Part I of this paper contained a technological description and economic evaluation of a large-scale highend CDW recycling plant in the Lisbon Metropolitan area. It concludes that economic viability is likely under the operating conditions considered, but these may and will very probably change in the near future. The reasons for such assumption have to do with the inherent uncertainty related to CDW generation (which might vary, for instance, due to socio-economic conditions in the region), such as the variability of CDW input gate fees and tariffs associated with landfilling rejected materials, which are market dependent parameters. This made it necessary to perform an (simplified) economic viability sensitivity analysis, focused on the investment return period and global economic balance. If parameters such as the plant’s capacity, the CDW input gate fee and landfill fee are varied, the investment return period is affected in different ways, though its value is generally kept below 8 years, for parameter variations of 30%. The analysis indicates economic performance for variations in single parameters, except for the plant’s capacity, which was considered to vary simultaneously with all others. Extreme best and worst scenarios were also tested in an attempt to define the model’s boundaries. Ó 2012 Elsevier Ltd. All rights reserved. Keywords: CDW fixed recycling plant Economic analysis Sensitivity analysis 1. Introduction Sensitivity analyses have proved to be extremely useful in scientific studies not only on building analysis (Junnila, 2004; International Energy Agency, 2005; Palme et al., 2008), but also on industrial products (Vadde et al., 2007) and CDW recycling operations (Bohne et al., 2008). Although special purpose methods have been developed to perform sensitivity analysis from a general mathematical standpoint, as in Saltelli et al. (2004), techniques such as variance-based methods and Monte Carlo filtering are not used in the present analysis, which as a consequence is greatly simplified. Very simple sensitivity analyses are performed regularly in studies which do not require complex probabilistic reasoning or to which it cannot be applied, because only basic linear parameter variation impacts are needed, very small sample sizes are available, or both (de Brito and Gonçalves, 2002; Saari, 2000; Dantana et al., 2004). This has been the case of the present study, since it was constructed on a simple spreadsheet environment, not compatible with the Monte Carlo filtering method, which involves: the use of available information or data to establish the set of acceptable model behaviours; in Monte Carlo * Corresponding author. Tel.: þ351 218443659; fax: þ351 218443071. E-mail addresses: jbrito@civil.ist.utl.pt, jb@civil.ist.utl.pt (J. de Brito). 0959-6526/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.jclepro.2012.05.006 methods parameters randomly vary over a range of values and generate corresponding sets of model predictions and classification of each simulation as acceptable or unacceptable according to pre specified behaviour definition (Rose et al., 1991). The application of this method would therefore require an automatic simulation engine, which would perform a set of algorithms (based on the CDW operation procedures) with randomly varied input parameters. This would greatly increase the complexity of the analysis, without introducing an equivalent benefit in the output results. The simplified sensitivity analysis used here derived from the need to see how varying the operating parameters would influence the return on investment period (and the 60-year operation period overall economic balance) for the CDW recycling plant studied (described and characterized in part I (Coelho and de Brito, submitted for publication)). The parameters in question are: the plant’s capacity, CDW input gate fee, concrete aggregate selling price, rejected materials landfill price, percentage of mixed/separated CDW input and CDW input mass rate. The parameters were varied separately, except for plant capacity, which was paired with each of the other parameters in turn. General simultaneous multi-parameter variations and couplings were not analysed as this would increase analysis time and complexity, and was not considered crucial. The only simultaneous parameter variation performed was conducted for two 330 A. Coelho, J. de Brito / Journal of Cleaner Production 39 (2013) 329e337 extreme scenarios of best and worst conditions. These two scenarios were tested for each facility capacity, but whilst it is highly unlikely that they will occur in the real facility they can be considered as boundaries to support the study’s framework. 2. Economic viability methodologyevariation parameters 2.1. Variation parameters As stated in the introduction, several variation parameters were chosen in order to evaluate the way investment return period and 60-year global economic balance were affected by these variations. These parameters were chosen initially for their perceived impact on the resultant performance factors which, except for the concrete aggregate selling price, were all proven to have moderate to high impact on the final results. However, other variables could have been studied, such as energy (electricity and diesel fuel) prices, the land acquisition cost (even though it is only about 1% of overall 60year costs) or the cost of purchasing and installing equipment (roughly 3% of overall 60-year costs). 2.1.1. Plant capacity Plant capacity, in tonne/h, was considered the main variation parameter, which means that all other variations in other parameters were performed in tandem with it. This was mainly to account, in a simplified way, for the uncertainty in CDW generation, in spite of its tendency to rise in the next decade or so (Rose et al., 1991). The plant was considered to be designed, depending on the actual CDW generation rate, for the following: 350 tonne/h, 250 tonne/h, 170 tonne/h, and 85 tonne/h. The base case analysed in part I has capacity of 350 tonne/h, derived from a 416 kg/person.year CDW generation rate applied to the Lisbon Metropolitan area (Coelho and de Brito, submitted for publication). 2.1.2. CDW input gate fee Part I showed that the CDW input gate fee benefits would be 86% of all benefits for a 350 tonne/h capacity plant. This striking figure naturally makes this parameter a potentially important factor in determining the plant’s economic viability. For the sensitivity analysis it was considered to vary by as much as 30%, over or under the calculated value of the average input gate fee for mixed CDW, V48/tonne, or V7.8/tonne for separated ceramic and concrete aggregates (Coelho and de Brito, submitted for publication). 2.1.3. Concrete aggregate selling price Recycled concrete aggregates comprise around 41% of all input mass and represent, unlike ceramics, a potential benefit in sales, with an average price of V2.8/tonne. Recycled ceramics have no market value at the moment, which is to say they can currently be delivered free of charge to recyclers/producers (Mimoso, 2011). Moreover, concrete aggregate represents 15% of all benefits from materials sold; but it only represents 2.3% of total benefits, which probably means that any change in the selling price of recycled concrete aggregate will not greatly affect the plant’s overall return on investment period. 2.1.4. Rejected materials landfill fee The average landfill fee was estimated at V114/tonne (taking figures ranging from V90/tonne to V150/tonne), derived from a market survey of regional waste operators and contractors. As concluded in part I, the overall share of landfill fees to be paid (assumed the same even if the material is delivered to downstream processors) over the 60 years operation period, is around 80% of all costs, which makes this parameter a prime candidate for sensitivity analysis. 2.1.5. Percentage of mixed/separated CDW input The plant operation was divided into two modes: one accepting and separating mixed CDW, and another only working with previously separated concrete and ceramic aggregates. The full operation mode involves higher operating costs than the simplified mode, in a proportion that varies according to the plant’s capacity, ranging from 4.7 through 2.9 more, respectively for case 1 (350 tonne/h capacity) and case 4 (85 tonne/h capacity). On the other hand, treating fully mixed CDW offers the possibility of charging a gate fee 6 higher than that for separated concrete and ceramic aggregates, with which comes large revenue. A third consequence of varying the percentage of mixed/separated CDW input is related to the rejected materials landfill fee, since the more mixed material treated, the more mass is rejected, and therefore higher costs must be supported. 2.1.6. CDW input mass CDW input mass is one of the most unpredictable factors considered and it could have a threefold negative impact, if reduced. For a given capacity design, a lower average input CDW flow will lead to sub-optimal operation, implying extra fixed installation costs not justified by the incoming CDW mass; moreover, lower CDW input, especially mixed CDW, translates into less benefits from gate fees and, finally, it also means lower output of materials to be sold. Assuming that a plant’s capacity cannot be exceeded (or that no surplus capacity can be obtained once the design has been established), then only reductions in CDW input flow matter to the analysis, since any incoming input material does not enter the plant if it is already running at full capacity. Consequently, reductions of 15 and 30% were considered, similar to other parameter variations. But assuming that this parameter has a particular effect on profitability and that it is subject to higher levels of uncertainty than the other variation parameters, a further reduction of 50% was considered in the analysis. 2.2. Case composition As noted in 2.1.1, the plant’s capacity was a variation parameter in all the sensitivity analyses performed. As such, for each design capacity all other parameters were considered to vary in turn, which renders this a two-dimensional sensitivity analysis. Table 1 shows the variation for each parameter and labelling each resulting case accordingly; each column’s variation results will therefore lead to three-dimensional charts, with the investment return period and global economic balance figures as their vertical axes. 3. Economic viability results and discussion - sensitivity analysis Table 2 and Table 3 summarise the main results obtained from the simplified sensitivity analysis, for each variation parameter. The investment return period results are given in Table 2 and the global economic balance over a 60-year period is presented in Table 3, to depict the potential payback or profit possible within that timeframe for the cases and variations analysed. The 3D charts produced from these tables are given in Figs. 1 and 2 through Fig. 10. Fig. 11 is given as an example of the time variation of the overall economic balance for the four plant capacities considered. It can immediately be gathered that installed capacity is of primordial importance to profitability and the return on investment period. It can be seen that capacity variation alone is responsible for doubling the return on investment period (2 years for a 350 tonne/h facility and 4 years for an 85 tonne/h facility), and an almost 6-fold reduction in the global 60-year economic balance (V386 million for a 350 tonne/h facility and V66.5 million for an 85 tonne/h facility). A. Coelho, J. de Brito / Journal of Cleaner Production 39 (2013) 329e337 331 Table 1 Case composition for sensitivity analysis. Plant capacity e main variation parameter Case 1e350 tonne/h Case 2e250 tonne/h Case 3e170 tonne/h Case 4e85 tonne/h Variation parameter CDW input gate fee Concrete aggregates selling price Rejected materials landfill fee % Of mixed/separated CDW input CDW input mass Case designation Case designation Case designation Case designation Case designation % Variation in relation to initial value Case 1_S4 Case 1_S3 Case 1_S0 Case 1_S1 Case 1_S2 Case 2_S4 Case 2_S3 Case 2_S0 Case 2_S1 Case 2_S2 Case 3_S4 Case 3_S3 Case 3_S0 Case 3_S1 Case 3_S2 Case 4_S4 Case 4_S3 Case 4_S0 Case 4_S1 Case 4_S2 30 15 0 15 30 30 15 0 15 30 30 15 0 15 30 30 15 0 15 30 % Variation in relation to initial value Case 1_T4 Case 1_T3 Case 1_T0 Case 1_T1 Case 1_T2 Case 2_T4 Case 2_T3 Case 2_T0 Case 2_T1 Case 2_T2 Case 3_T4 Case 3_T3 Case 3_T0 Case 3_T1 Case 3_T2 Case 4_T4 Case 4_T3 Case 4_T0 Case 4_T1 Case 4_T2 30 15 0 15 30 30 15 0 15 30 30 15 0 15 30 30 15 0 15 30 Case 1_U4 Case 1_U3 Case 1_U0 Case 1_U1 Case 1_U2 Case 2_U4 Case 2_U3 Case 2_U0 Case 2_U1 Case 2_U2 Case 3_U4 Case 3_U3 Case 3_U0 Case 3_U1 Case 3_U2 Case 4_U4 Case 4_U3 Case 4_U0 Case 4_U1 Case 4_U2 3.1. Variations in selected parameters As shown in Fig. 1, the CDW input gate fee has a pronounced effect on the investment return period for any installed capacity (especially for lower capacities, though). For a 30eþ30% variation in the CDW input gate fee the return period varies by a minimum factor of 2.5 (170 tonne/h facility) and a maximum of 4 (85 tonne/h facility). To guarantee high profitability and avoid unattractive return periods, gate fees must be kept as high as possible, while competiveness with landfills and other waste processors is maintained. If plant installed capacity is 85 tonne/h, return on investment periods may become too long (more than the 8 years taken as % Variation in relation to initial value 30 15 0 15 30 30 15 0 15 30 30 15 0 15 30 30 15 0 15 30 Case Case Case Case Case Case Case Case Case Case Case Case Case Case Case Case Case Case Case Case 1_V4 1_V3 1_V0 1_V1 1_V2 2_V4 2_V3 2_V0 2_V1 2_V2 3_V4 3_V3 3_V0 3_V1 3_V2 4_V4 4_V3 4_V0 4_V1 4_V2 % Variation in relation to initial value 30 15 0 15 30 30 15 0 15 30 30 15 0 15 30 30 15 0 15 30 Case Case Case Case e Case Case Case Case e Case Case Case Case e Case Case Case Case e % Variation in relation to initial value 1_X0 1_X1 1_X2 1_X3 0 15 30 50 e 2_X0 2_X1 2_X2 2_X3 0 15 30 50 e 3_X0 3_X1 3_X2 3_X3 0 15 30 50 e 4_X0 4_X1 4_X2 4_X3 0 15 30 50 e a reference for economic viability in some regulations (Coelho and de Brito, submitted for publication)), especially if possible gate fees remain 15e30% below the averages considered here. This, however, will depend on how initial fundraising is dealt with and negotiated. Global economic balance is also strongly affected by the CDW input gate fee. For a 350 tonne/h facility, a 30% decrease in its value over 60 years represents a 63% reduction in global economic balance. That decrease is even sharper for an 85 tonne/h facility, which ends up with a 67% fall in global economic balance. Fig. 3 shows that concrete aggregate selling price (coarse or fine, within a 30% variation range) does not have much influence on the return on investment period, for any of the capacities evaluated. Accordingly, influence on the global economic balance over 60 years Table 2 Results for investment return period, for all analysed situations. Plant capacity e main variation parameter Case 1e350 tonne/h Case 2e250 tonne/h Case 3e170 tonne/h Case 4e85 tonne/h Variation parameter CDW input gate fee Concrete aggregates selling price Rejected materials landfill fee % Of mixed/separated CDW input CDW input mass Case designation Value, years Case designation Value, years Case designation Value, years Case designation Value, years Case designation Value, years Case 1_S4 Case 1_S3 Case 1_S0 Case 1_S1 Case 1_S2 Case 2_S4 Case 2_S3 Case 2_S0 Case 2_S1 Case 2_S2 Case 3_S4 Case 3_S3 Case 3_S0 Case 3_S1 Case 3_S2 Case 4_S4 Case 4_S3 Case 4_S0 Case 4_S1 Case 4_S2 1 1 2 2 3 1 1 2 2 3 2 2 2 3 4 3 3 4 5 9 Case 1_T4 Case 1_T3 Case 1_T0 Case 1_T1 Case 1_T2 Case 2_T4 Case 2_T3 Case 2_T0 Case 2_T1 Case 2_T2 Case 3_T4 Case 3_T3 Case 3_T0 Case 3_T1 Case 3_T2 Case 4_T4 Case 4_T3 Case 4_T0 Case 4_T1 Case 4_T2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 4 4 4 4 4 Case 1_U4 Case 1_U3 Case 1_U0 Case 1_U1 Case 1_U2 Case 2_U4 Case 2_U3 Case 2_U0 Case 2_U1 Case 2_U2 Case 3_U4 Case 3_U3 Case 3_U0 Case 3_U1 Case 3_U2 Case 4_U4 Case 4_U3 Case 4_U0 Case 4_U1 Case 4_U2 2 2 2 1 1 2 2 2 2 1 3 2 2 2 2 6 4 4 3 3 Case 1_V4 Case 1_V3 Case 1_V0 Case 1_V1 Case 1_V2 Case 2_V4 Case 2_V3 Case 2_V0 Case 2_V1 Case 2_V2 Case 3_V4 Case 3_V3 Case 3_V0 Case 3_V1 Case 3_V2 Case 4_V4 Case 4_V3 Case 4_V0 Case 4_V1 Case 4_V2 1 1 2 2 2 1 2 2 2 2 2 2 2 3 3 2 3 4 5 6 Case 1_X0 Case 1_X1 Case 1_X2 Case 1_X3 e Case 2_X0 Case 2_X1 Case 2_X2 Case 2_X3 e Case 3_X0 Case 3_X1 Case 3_X2 Case 3_X3 e Case 4_X0 Case 4_X1 Case 4_X2 Case 4_X3 e 2 2 2 2 e 2 2 2 3 e 2 2 3 4 e 4 5 6 10 e 332 A. Coelho, J. de Brito / Journal of Cleaner Production 39 (2013) 329e337 Table 3 Results for the global economic balance over 60 years, for all analysed situations. Plant capacity e Variation parameter main variation CDW input gate fee Concrete aggregates selling price Rejected materials landfill fee % Of mixed/separated CDW input CDW input mass parameter Case Economic Case Economic Case Economic Case Economic Case Economic balance over designation balance over designation balance over designation balance over designation balance over designation 60 years, 60 years, 60 years, 60 years, 60 years, V  106 V  106 V  106 V  106 V  106 Case 1e350 tonne/h Case 2e250 tonne/h Case 3e170 tonne/h Case 4e85 tonne/h Case 1_S4 Case 1_S3 Case 1_S0 Case 1_S1 Case 1_S2 Case 2_S4 Case 2_S3 Case 2_S0 Case 2_S1 Case 2_S2 Case 3_S4 Case 3_S3 Case 3_S0 Case 3_S1 Case 3_S2 Case 4_S4 Case 4_S3 Case 4_S0 Case 4_S1 Case 4_S2 591 489 386 284 145 416 343 270 197 123.4 275 225 175 126 75.8 125 99.7 74.8 49.9 25.0 Case 1_T4 Case 1_T3 Case 1_T0 Case 1_T1 Case 1_T2 Case 2_T4 Case 2_T3 Case 2_T0 Case 2_T1 Case 2_T2 Case 3_T4 Case 3_T3 Case 3_T0 Case 3_T1 Case 3_T2 Case 4_T4 Case 4_T3 Case 4_T0 Case 4_T1 Case 4_T2 391 389 386 384 382 273 272 270 268 266 178 177 175 174 173 76.0 75.4 74.8 74.2 73.6 Case 1_U4 Case 1_U3 Case 1_U0 Case 1_U1 Case 1_U2 Case 2_U4 Case 2_U3 Case 2_U0 Case 2_U1 Case 2_U2 Case 3_U4 Case 3_U3 Case 3_U0 Case 3_U1 Case 3_U2 Case 4_U4 Case 4_U3 Case 4_U0 Case 4_U1 Case 4_U2 is also negligible (Fig. 4), as varying this parameter by 30% will not change it by more than 1.6%, for any of the installed capacities. The landfill fee for rejected materials has a moderate influence on the return on investment and global economic balance, as is apparent from Figs. 5 and 6. As shown, the investment return period never exceeds 5 years for any installed capacity, even in particularly adverse conditions (a 30% increase in landfill fees when operating an 85 tonne/h plant). In fact, variations in the investment return period never exceed 50% for the 30% variation in the CDW landfill fee. 294 340 386 432 478 204 237 270 303 336 131 153 175 198 220 52.4 63.6 74.8 86.0 97.2 Case 1_V4 Case 1_V3 Case 1_V0 Case 1_V1 Case 1_V2 Case 2_V4 Case 2_V3 Case 2_V0 Case 2_V1 Case 2_V2 Case 3_V4 Case 3_V3 Case 3_V0 Case 3_V1 Case 3_V2 Case 4_V4 Case 4_V3 Case 4_V0 Case 4_V1 Case 4_V2 604 485 386 306 249 422 338 270 213 170 281 224 175 137 107.4 127 98.4 74.3 55.0 40.4 Case 1_X0 Case 1_X1 Case 1_X2 Case 1_X3 e Case 2_X0 Case 2_X1 Case 2_X2 Case 2_X3 e Case 3_X0 Case 3_X1 Case 3_X2 Case 3_X3 e Case 4_X0 Case 4_X1 Case 4_X2 Case 4_X3 e 386 322 260 175 e 270 224 178 116 e 175 144 113 70.8 e 74.3 58.6 42.8 21.7 e Meanwhile, differences in the global economic balance for the cases considered are consistent with the parameter variation range. So a 350 tonne/h capacity plant would see a 30% reduction in global economic balance for a 30% reduction in the landfill fee for rejected materials; for an 85 tonne/h plant this relation is somewhat different, but still only gives a change of 24% in the 60 years global economic balance. However, to reduce sensitivity to this factor while maintaining high profitability, higher installed capacities must be sought (more than 170 tonne/h and preferably in the 250e350 tonne/h range). Fig. 1. Return on investment period, years e CDW input gate fee. A. Coelho, J. de Brito / Journal of Cleaner Production 39 (2013) 329e337 333 Fig. 2. Global economic balance over a 60-year period e CDW input gate fee. Fig. 4. Global economic balance over a 60-year period e concrete aggregates selling price. As for the percentage of mixed/separated CDW input, Figs. 7 and 8 show that the more mixed CDW materials enter the facility the higher the global economic balance (the more benefits) and the lower the return on investment period. In fact, except for the 85 tonne/h capacity case, all return on investment periods remain comfortably below 4 years, even if mixed CDW input gets reduced to about 50% of the installed capacity (the rest of the input being composed of separated aggregate). This result is directly linked with the CDW input gate fee, from which the facility can of course generate more income if it can charge the higher amount associated with mixed CDW input, as often as possible. This parameter has a mild influence on the return on investment period, when variations of 30% never generate differences of more than 50% in the result. The 85 tonne/h facility is most sensitive to it, however, since the return on investment period changes by a factor of 3 for a 30% variation, while in the 170 tonne/h plant it is a factor of 1.5 (with the other two capacities exhibiting change factors of 2, for the same parameter variation range). Global economic balance is rather more affected by this parameter, especially for larger amounts of mixed CDW materials. Actually, if þ30% of the latter is input (reaching 91% of all input CDW), the global economic balance increased by 71%, compared with the unchanged situation (70% mixed and 30% separated CDW), for the 85 tonne/h facility, and 56% for the 350 tonne/h one. Fig. 3. Return on investment period, years e concrete aggregates selling price. When the parameter is reduced, however, the response is not symmetrical, as Fig. 8 shows. In this case a 30% reduction in mixed CDW input implies a 36% change in global economic balance for the 350 tonne/h unit and a 46% change in the 85 tonne/h capacity one. This asymmetry is related to the cumulative effect of changing the input mix: more mixed CDW input entails more benefits gained due to the higher input rate than benefits lost if the lower rate is applied more often; also, more mixed CDW input will imply, although to a lesser degree, higher benefits from output material sales than benefits lost if it is reduced. This occurs in spite of the fact that energy, labour and rejected materials dumping costs are higher when a greater mixed CDW flow enters the facility than when it is reduced. Analysing varying CDW input mass, from Fig. 9 it can be seen that the return on investment period is affected, but especially so for the 170 and 85 tonne/h installed capacities. For these last two, reducing the CDW input down to 50% of the designed capacity can aggravate the return on investment period by a factor of 2 and 2.5, respectively, tending towards prohibitive values for the 85 tonne/h plant. Global economic balance shows an even clearer outcome, since it is reduced by a factor of 2.2 for the 350 tonne/h plant, by a factor of 3.4 for the 85 tonne/h one. Once again it becomes clear Fig. 5. Return on investment period, years e rejected materials landfill fee. 334 A. Coelho, J. de Brito / Journal of Cleaner Production 39 (2013) 329e337 Fig. 6. Global economic balance over a 60-year period e rejected materials landfill fee. Fig. 7. Return on investment period, years e percentage of mixed/separated CDW input. A. Coelho, J. de Brito / Journal of Cleaner Production 39 (2013) 329e337 Fig. 8. Global economic balance over 60-year period e percentage of mixed/separated CDW input. that higher capacity outperform lower capacity facilities, both in absolute profitability and in resilience to the fluctuations of several operating parameters (this is in tune with other studies, such as Zhao et al. (2010) and Duran et al. (2006)). 3.2. Best and worst scenarios Best and worst scenarios were tested for each installed capacity by assuming all parameters respectively at their most and least favourable values. Specifically, parameter favourable conditions are:     CDW input gate fee: þ30%; Rejected materials landfill fee: 30%; Percentage of mixed/separated CDW input: þ30%; CDW input mass: equal to the plant’s capacity. The concrete aggregate selling price was ignored since, as seen above, it has a negligible influence on results. Worst parameter conditions will be the opposite, with a 50% factor applied to the Fig. 9. Return on investment period, years e CDW input mass. 335 Fig. 10. Global economic balance over 60-year period e CDW input mass. CDW input mass. Results are summarized in Table 4, where, for each plant capacity and these best and worst scenarios, the 60-year global economic balance and investment return period are given. The time dependent overall economic balance is given in Fig. 12 (best scenario) and 13 (worst scenario), for all the capacities. These scenarios were tested to determine the facility’s model behaviour to extreme, though extremely unlikely, conditions. It can be seen from these results that changing all parameters simultaneously, in a favourable or unfavourable direction, has a profound effect on economic results. This “group” variation, when unfavourable, prevents the lower capacity facilities e 85 tonne/h and 170 tonne/h e from being economically viable. Even the 250 tonne/h facility, in its worst scenario, can barely be seen as viable, with an intolerable return on investment period of 28 years. Changing all specified parameters in an unfavourable direction sharply reduces operating benefits, with a halving of CDW input mass, a 30% cut in the input CDW gate fee, halving of the recycled output mass (i.e. less revenue from selling materials) and less mixed input CDW, which limits CDW mass chargeable at higher prices (mixed CDW). At the same time, this worst scenario implies higher costs from dumping fees, which suffer a 30% increase. As a result, only the 350 tonne/h facility has a chance of economic viability (even though 14 years of return on investment period will probably raise concern among investors), in these theoretical worst cases. Fig. 11. Overall economic balance for the 60-year operation period for four different plant capacities. (350, 250, 170 and 85 tonne/h). 336 A. Coelho, J. de Brito / Journal of Cleaner Production 39 (2013) 329e337 Table 4 Best and worst scenarios for each facility capacity. Plant’s capacity Case designation Economic balance over 60 years, MV Investment return period, years Case 1e350 tonne/h Case Case Case Case Case Case Case Case 1081 41.7 544 6.2 513 3.7 168 14.7 1 8 1 28 1 No return 2 No return Case 2e250 tonne/h Case 3e170 tonne/h Case 4e85 tonne/h 1_Best 1_Worst 1_Best 1_Worst 1_Best 1_Worst 1_Best 1_Worst Notes. 1. Negative values represent benefits. 2. "No return" means permanent global economic balance deficit or zero return on investment period over 60 years. Fig. 13. Overall economic balance for the 60-year operation period for four different plant capacities e worst scenario. Table 4 also makes it clear that, where the 60-year overall economic balance is still positive, with net benefits to the owner, the resulting value changes by as much as 99% from the best to the worst case scenarios. In fact, for the 350 tonne/h and 250 tonne/h facilities, after 60 years of operation global benefits are reduced by 89% and global costs by 67% (decrease in costs due to less material being rejected e less paid out in dumping fees e and more operating time in the simplified mode, which represents less energy and labour costs). This asymmetry means that, in the worst case scenario, global benefits and costs will be closer than in the best case scenario, which implies a further reduction in the global economic balance (in this case, 99%). As a consequence, the overall economic balance is reduced by roughly two orders of magnitude, from best to worst case scenarios (Figs. 12 and 13). In the latter case, Fig. 13 also shows that certain investment moments, particularly after 20 and 40 years of operation, are much clearer than in best case scenarios (or even than in the base case scenario, where parameters are at their original values e Fig. 11). Due to the sharp reduction in global economic balance in the worst scenario, the magnitude of the investment required for equipment replacement, especially at the stated moments, get closer to the best scenario (which generates the serrated appearance in the global economic balance curves). As operating conditions deteriorate, therefore (economically relevant parameters worsen), it becomes more important to pay attention to equipment maintenance and conservation. If, by investing more time and money in extending all the equipment’s service life, replacement periods get larger, then that Fig. 12. Overall economic balance for the 60-year operation period for four different plant capacities - best scenario. will be financially more important in unfavourable conditions than in regular or favourable conditions. On the best case scenarios side, return on investment period never rises above 2 years, for all capacities analysed. This is basically due to higher benefits as more mixed CDW enter the facility and a higher gate fee is charged, but also to reduce costs, as the dumping fee falls. Of course these conditions are excellent for the investor, but they are considered unlikely as all parameters reach best values simultaneously. 4. Conclusions Following the analysis described in Part I, here we have explained the economic sensitivity analysis performed on an advanced CDW recycling plant located in the Lisbon Metropolitan area, registering the results on the return of investment period and the global economic balance over a 60-year operation period, given by varying six parameters: the plant’s capacity (tonne/h), CDW input gate fee (V/tonne), concrete aggregate sales price (V/tonne), rejected materials landfill fee (V/tonne), percentage of mixed/ separated CDW input (%) and CDW input mass (tonne). Also a duo of extreme best and worst case scenarios was tested, in order to evaluate the facility’s economic performance in those conditions and establish framework limits. The following conclusions can be drawn:  In general, return on investment period were under 8 years (1 or 2 years in favourable conditions), for a wide range of parameter variations. In fact, this figure is exceeded in only two particularly aggravated conditions e the 85 tonne/h plant, functioning with a 30% cut in CDW input fee, and (separately) the 85 tonne/h plant in which CDW input mass is reduced by 50%. Given the generally higher return on investment periods found for the 85 tonne/h capacity plant and its higher variability (compared with the same analysis for other plant capacities), this result was expected. As a consequence, in order to guarantee return periods of less than 8 years and less sensitivity to changing operating parameters, design capacities under 170 tonne/h should be avoided;  The most favourable operating conditions occur, within the range considered and for single parameter variations, when the CDW input fee is 30% above the average value determined or the percentage of mixed/separated CDW input is as close to completely mixed as possible. Both these parameters influence A. Coelho, J. de Brito / Journal of Cleaner Production 39 (2013) 329e337      the result equally, as they maximize the benefit from charging the highest possible CDW input fee; consequently, this fee must be given top priority and be managed as accurately as possible (given particular local market conditions), since it has a strong effect on the facility’s profitability; The global economic balance for a 60-year operation period is always beneficial to the owner/manager; given the cost-benefit conditions of all the analysed situations, even in the worst scenario a beneficial global outcome is expected after 12 years of operation. Gains could be as high as V555 million (350 tonne/h capacity plant, operating at þ30% of the average CDW input gate fees) and as low as V16.6 million (85 tonne/h capacity plant, operating at 30% of the average CDW input gate fees). This is valid for single variation parameters, for all capacities simulated; Economic viability for a full-scale high-end CDW recycling plant is likely to occur for a widely varying operating range of economic parameters, in pure open market conditions, i.e., without government support or specific legislation in favour of recycling CDW; Although specific parameter coupling influences, e.g. low CDW mass input (relative to plant capacity) and lower than average CDW input gate fee, were not analysed in detail (this kind of multi-parameter sensitivity analysis can be the object of further research), extreme conditions of parameters that vary simultaneously in favourable and unfavourable directions do have a pronounced effect on economic performance. Best conditions can lower the investment return period for any of the capacities studied to under 2 years, and worst case conditions turn smaller facilities (85 and 170 tonne/h) into unprofitable units, and jeopardise the profitability of larger facilities (250 and 350 tonne/h); In particularly adverse conditions e 50% less input CDW mass, 30% more separated CDW entering the facility, 30% lower input CDW gate fee and 30% higher dumping costs e investment in maintenance and conservation becomes particularly important, as postponing the replacement of equipment can have significant financial consequences and shorten the return on investment period or increase the overall economic balance over the years; The favourable economic outcome demonstrated by this study must be strictly related to the conditions implied, which might benefit from lack of competition (just one large-scale facility 337 for the entire region) and assumption of static conditions over the years (not accounting for dynamic variations over time). Acknowledgements Thanks are due to the FCT (Foundation for Science and Technology) for the research grant awarded to the first author and to the ICIST e IST research centre. References Bohne, R.A., Brattebø, H., Bergsdal, H., 2008. Dynamic eco-efficiency projections for construction and demolition waste recycling strategies at the city level. Journal of Industrial Ecology 12 (1), 52e66. Coelho, A., de Brito, J., submitted for publication. Economic viability of a construction and demolition waste recycling plant in Portugal e part I: location, materials, technology and economic analysis. Journal of Cleaner Production. Dantana, N., Touran, A., Wang, J., 2004. An analysis of cost and duration for deconstruction and demolition of residential buildings in Massachusetts. Resources, Conservation and Recycling 44 (1), 1e15. de Brito, J., Gonçalves, A.P., 2002. Economic viability analysis of concrete aggregate recycling (in Portuguese). In: Concrete for Sustainable Construction Seminar, Lisbon, Portugal. Duran, X., Lenihan, H., O’Regan, B., 2006. A model for assessing the economic viability of construction and demolition waste recycling e the case of Ireland. Resources, Conservation and Recycling 46 (3), 302e320. International Energy Agency, 2005. IEA Annex 31. Energy related environmental impacts of buildings. http://www.iisbe.org/annex31/index.html. Junnila, S., 2004. The environmental impact of an office building throughout its life cycle. Doctoral dissertation, Research Report 2, Helsinki University of Technology Construction Economics and Management, Espoo, Finland. Extended overview: http://lib.tkk.fi/Diss/2004/isbn9512272857/isbn9512272857.pdf. Mimoso, P., 2011. Personal Communication. Visa Consultants, Lisbon, Portugal. January, 2011. Palme, M., Isalgué, A., Coch, H., Serra, R., 2008. Building sensitivity to climatic fluctuations and user’s actions: a challenge for high-tech buildings. In: PLEA 2008-25th Conference on Passive and Low Energy Architecture, Dublin, 22nde24th October, 2008. Rose, K.A., Smith, E.P., Gardner, R.H., Brenkert, A.L., Bartell, S.M., 1991. Parameter sensitivities, Monte Carlo filtering, and model forecasting under uncertainty. Journal of Forecasting 10 (1e2), 117e133. Saari, A., 2000. Management of life-cycle costs and environmental impacts in building construction. In: RILEM/CIB/ISO International Symposium, vol. 14. Helsinki, Finland, pp. 117e122. Saltelli, A., Tarantola, S., Campolongo, F., Ratto, M., 2004. Sensitivity Analysis in Practice e a Guide to Accessing Scientific Models. Joint Research Centre of the European Commission, Ispra, Italy. Vadde, S., Kamarthi, S.V., Gupta, S.M., 2007. Optimal pricing of reusable and recyclable components under alternative product acquisition mechanisms. International Journal of Product Research 45 (18), 4621e4652. Zhao, W., Leeftink, R.B., Rotter, V.S., 2010. Evaluation of the economic feasibility for the recycling of construction and demolition waste in China e The case of Chongqing. Resources, Conservation and Recycling 54 (6), 377e389.