### Abstract

- Top of page
- Abstract
- 1. EVOLUTION OF THE CONCEPT OF CULTURAL HERITAGE
- 2. THE SUSTAINABILITY OF THE RE-USE OF HISTORIC ARTEFACTS
- 3. MULTICRITERIA AGGREGATION
- 4. EVALUATION OF SUSTAINABILITY OF RE-USE PROJECTS
- 5. EVALUATION OF SUSTAINABILITY OF HYPOTHESIS FOR THE RE-USE OF THE HISTORIC VENETIAN ARSENAL
- 6. THE ASSESSMENT OF SUSTAINABILITY OF RE-USE PROJECTS
- 7. CONCLUSIONS
- Acknowledgements
- REFERENCES

The paper presents a multiple criteria model for the evaluation of the sustainability of projects for the economic re-use of historical buildings in Venice. The model utilizes the relevant parameters for the appraisal of sustainability, aggregated into three macro-indicators: intrinsic sustainability, context sustainability and economic-financial feasibility. The model has been calibrated by a panel of experts and tested on two reuse hypothesis of the Old Arsenal in Venice. Copyright © 2011 John Wiley & Sons, Ltd.

### 1. EVOLUTION OF THE CONCEPT OF CULTURAL HERITAGE

- Top of page
- Abstract
- 1. EVOLUTION OF THE CONCEPT OF CULTURAL HERITAGE
- 2. THE SUSTAINABILITY OF THE RE-USE OF HISTORIC ARTEFACTS
- 3. MULTICRITERIA AGGREGATION
- 4. EVALUATION OF SUSTAINABILITY OF RE-USE PROJECTS
- 5. EVALUATION OF SUSTAINABILITY OF HYPOTHESIS FOR THE RE-USE OF THE HISTORIC VENETIAN ARSENAL
- 6. THE ASSESSMENT OF SUSTAINABILITY OF RE-USE PROJECTS
- 7. CONCLUSIONS
- Acknowledgements
- REFERENCES

Mapping out the guidelines for the sustainable economic re-use of historic buildings cannot leave out of consideration the complexity of the objectives and methodologies for safeguarding cultural heritage. The historic, aesthetic and artistic characteristics of cultural assets make it difficult to apply a solely qualitative approach. The complexity of the investigation is also due to the public nature of these goods, not necessarily as far as the property right is concerned—many are privately owned—but rather to those related to historic, artistic and cultural values [Brosio, 1993]. The evolution of the concept of cultural heritage in Italian laws and regulations is very interesting. An important law for this matter, passed in 19391, deals with ‘moveable and immovable assets which are of artistic, historic, archaeological or ethnographic interest’, considered being objects which are aesthetically pleasing and, as such, should be safeguarded by appropriate legislation. Article 9 of the Constitutional Charter refers to these concepts and states: ‘The Republic […..] safeguards natural landscape and the historical and artistic heritage of the Nation’, affirming the central Government's sovereignty over the cultural heritage and the values of national identity [Giannini, 1976].

Italy's post-war cultural debate developed new views by proposing innovative laws and Commissions, including the Franceschini2 Commission, which first used the term ‘cultural heritage’ to describe ‘material evidence of civil value’. The cultural heritage assets are no longer simply aesthetically pleasing but also a palimpsest of a culture's history.

The cultural heritage and landscape is currently safeguarded by the ‘Codex of cultural heritage’,3 which, together with the prior law,4 defines cultural heritage. According to this definition, cultural heritage are assets that also encompass the qualities and attributes of objects that have ethnic, anthropological, archivistic or literary value for past, present or future generations.

In recent years, there has been an increasing interest concerning the economic value of the cultural heritage, defining the economic value not only in monetary terms but also in terms of a broader considerations, recognizing, for instance, the fact that the conservation of these assets also generates economic benefits to the society as a whole [Forte, 1977; Throsby, 2002].

Throsby defines cultural heritage as ‘an asset which embodies, stores or provides cultural value in addition to whatever economic value it may possess’ [Throsby, 2001, 2002]. The difference, however, between physical assets (from a strictly economic viewpoint) and cultural capital is indeed the concept of ‘culture’, which bestows the historic goods with an added qualitative dimension. It is this cultural quality that must be maintained and not simply the materials with which the asset is built.

In the scientific literature [Randall, 1991; Stellin and Rosato, 1998], the economic and cultural value of a historic asset are to be distinguished in two macro-categories which refer to two spatial and temporal dimensions. The difference lies in the use and non-use value:

The use value, linked to the benefits the consumer receives directly from the asset itself, is a contingent prerogative; it is the utility that the historic artefact offers the consumer from the very moment he comes into contact with it. For this reason synchrony must be created between the cultural asset and the user;

The non-use value, instead, does not have the same contingent obligation of the above and, as a result, does not require such close synchrony (but rather a diachrony) as it refers to the utility that the consumers perceive from the conservation of the cultural assets for themselves and for the future generations.

### 2. THE SUSTAINABILITY OF THE RE-USE OF HISTORIC ARTEFACTS

- Top of page
- Abstract
- 1. EVOLUTION OF THE CONCEPT OF CULTURAL HERITAGE
- 2. THE SUSTAINABILITY OF THE RE-USE OF HISTORIC ARTEFACTS
- 3. MULTICRITERIA AGGREGATION
- 4. EVALUATION OF SUSTAINABILITY OF RE-USE PROJECTS
- 5. EVALUATION OF SUSTAINABILITY OF HYPOTHESIS FOR THE RE-USE OF THE HISTORIC VENETIAN ARSENAL
- 6. THE ASSESSMENT OF SUSTAINABILITY OF RE-USE PROJECTS
- 7. CONCLUSIONS
- Acknowledgements
- REFERENCES

The valuation of the sustainability of the economic re-use of historical heritage is crucial on this discussion and helps to tailor safeguard and protection policies.

Starting with the well-known declination of the concept of economic, social and environmental sustainability, literature on the matter refers to a common premise according to which the ultimate objective of any type of intervention should be to develop local resources and, as a consequence, to contribute to enhancing the quality of life. This is a multi-dimensional concept in so far as ‘the quality of life’ touches several different economic and social aspects [Fusco Girard, 1987; Howarth, 1997].

The concept of sustainability was initially presented by the World Commission on Environment and Development [1987] with reference to the effects of development on environmental assets. Sustainable development was defined as ‘development that meets the needs of the present without compromising the ability of future generations to meet their own needs’.

As far as the cultural heritage and, in particular, architectural assets are concerned, the concept of sustainability is influenced by the environment and involves two main aspects: the sustainability of the material and formal transformation of the building and the sustainability of the new function that is to be installed therein. In other words, the objective of sustainability requires an equilibrium between the economic re-use of the asset and its conservation [Nijkamp and Voogd, 1989].

Current debate on the theories of restoration philosophies, which is particularly active in Italy today, follows two lines of thought.

The first is defined as *critical restoration*, and stems from the conviction that each intervention project represents a case of its own. Restoration must also transmit the asset to the future by guaranteeing and facilitating its interpretation without loosing sight of the fact that it is a ‘non-verbal criticism expressed in concrete non-verbal ways’ [Carbonara, 1987; Marconi, 1993].

On the other hand, we have the pure *conservationists* who support the conservation of each strata of material or matter that the building has accumulated over time. Under this approach the building becomes a sort of palimpsest where it is impossible to identify what exactly has to be conserved or removed: ‘The aim of restoration is to conserve both the matter and substance which represents an archive of what the building is actually made of’ [Dezzi, 1991].

An economic re-use project, attributing a new function to the building, often involves transforming, consolidating, adding and removing and may alter the various strata of existing materials and structures.

The decision not to use an asset, however, undermines the intrinsic value of the asset and use poses the threat of possible abandonment and subsequent loss of the asset on the whole.5

Often, however, historic architectural complexes are used for purposes that are completely different from those for which they had originally been built and the interventions required (especially in terms of standards and building regulations that need to be respected) might not always be compatible with the typology and structure of the architectural asset on which works are being carried out. Over-use or incompatible use can have similar consequences to those of abandonment and can gradually reduce the cultural value and historic evidence of the artefact.

Literature does not deal with the definition of what is, or is not sustainable as far as work carried out on historic buildings are concerned. One of the reasons for this silence might be sought in interdisciplinary character of the issue. In this paper a multiple criteria valuation model is proposed which is able to tackle interdisciplinary problems of valuation. The model is founded on a set of parameters measuring the performances of the reuse project.

From the informations codified in parameters, a set of indicators can been developed representing the different points of view with which the concept of ‘sustainability’ may be implemented in the case of restoration and reuse of historic buildings. These indicators should gear to the aim of identifying the limit of transformations, helping to identify the point at which the new use ceases to enhance the asset, and begins to consume and erode the original value. In the following paragraph, the quantitative framework utilized to implement such a model is presented.

### 3. MULTICRITERIA AGGREGATION

- Top of page
- Abstract
- 1. EVOLUTION OF THE CONCEPT OF CULTURAL HERITAGE
- 2. THE SUSTAINABILITY OF THE RE-USE OF HISTORIC ARTEFACTS
- 3. MULTICRITERIA AGGREGATION
- 4. EVALUATION OF SUSTAINABILITY OF RE-USE PROJECTS
- 5. EVALUATION OF SUSTAINABILITY OF HYPOTHESIS FOR THE RE-USE OF THE HISTORIC VENETIAN ARSENAL
- 6. THE ASSESSMENT OF SUSTAINABILITY OF RE-USE PROJECTS
- 7. CONCLUSIONS
- Acknowledgements
- REFERENCES

Many methods have been proposed in the literature to approach multicriteria problems. Following Vincke *et al.* [1989], a commonly used classification distinguishes

Approaches derived from Multi Attribute Value Theory (MAVT);

Outranking approaches, like the ELECTRE family and its derivates;

The MAVT methods compute a score for each alternative, using Aggregation Operators (AO) [Klement *et al*., 2000; Kolesàrovà and Mordelovà, 2001]. Many of the MAVT methods are based on common sense rules, tailored for not quantitative skills of the majority of the Decision Makers (DM). In this contribution, we propose a mathematically founded MAVT approach, which is at the same time easily to be understood by any DM given a suitably designed interface.

The most common aggregation operator is the (simple) Weighted Averaging approach (WA),6 which, for each alternative, computes the weighted average of the criterion score. It is a simple and intuitive compensative method, but no interaction among the criteria can be admitted, since it is based on the Independent Preference axiom. For this reason, many other methods were proposed. We limit to quote the Geometric Averaging (GA) which computes the geometrical averaging of the criterion scores. It can be usefully applied in *strong* conservative cases, since it gives a null global score if at least one criterion is null (thus impeding compensation). Another class of Aggregation Operators consists of the Ordered Weighted Averaging operators (OWA) introduced by Yager [1988, 1992]. It includes, as particular cases, the weighted averaging, and, as extreme situations, the Max and the Min operators. If the weights are obtained by a non-monotonic quantifier [Yager, 1993], the OWA operator implements linguistic statements as ‘at least’, ‘at most’, ‘at least the half’ and so on. The *compensation operator* introduced by Zimmermann [Von Altrock, 1995] uses a tuning parameter, representing thus more or less conservative situations. A different approach is obtained using a Fuzzy Expert System, but its design is not a simple task, since many effort needs to be devoted to the inference rules definition [Von Altrock, 1995].

More recently, the introduction of methods based on non-additive measures (NAM) helped to solve many theoretically cumbersome problems, and at the same time offers a wide range of possibilities of aggregation. Up to now, the multicriteria community considers these methods the most complete and mathematically well-founded MAVT approach. Roughly speaking, NAM consists of assigning a suitable weight to *every possible* coalition of the state of the criteria, and not only to a *single* criterion, as the WA approach. So the importance of a coalition of criteria can be greater, equal or less than the sum of the importance (weights) of each criterion included in the coalition. Both *synergic* and *redundancy* interactions among the criteria can be modelled in this way. If the importance of the coalition for each of them is equal to the sum of weights of the included criteria, the operator simplifies to the WA approach. In the other cases, a simple algorithm computes the score of the alternatives, considering the interactions among the criteria given by the non-additive measures. Moreover, some indices can be computed showing the tendency towards pessimism or optimism reflected in the valuation of the set of alternatives. It should be remembered that the NAM can be directly obtained by experimental data, or implicitly elicited from expert's judgements. In this contribution, we propose an implicit approach. The price to be paid with respect to WA or OWA consists of an increase in the number of parameters, which are equal to the number of all possible coalitions of criteria. For example, using only two possible states for each criteria, 4 criteria request 16 parameters, with 5 criteria 32 parameters, and with 6 criteria 64 parameters are needed. Verifying the absence of interaction between higher order coalitions, we can use a *reduced order* model where the number of parameters is strongly reduced [Grabish, 1997].

#### 3.1. Non-additive measures

Let . A *non additive* measure, [Marichal, 1998, 1999a, b] is a set function , so that the following conditions hold:

Such a measure is able to represent interactions among the criteria, giving a different weight to every possible coalition of them, and not only to a single one as in the case of the WA operator. The first and the third conditions limit the variability inside the domain [0,1], while the second condition is a monotonicity constraint, namely, if more criteria are satisfied, the global satisfaction cannot decrease.7

A non additive measure will be named as:

For an additive measure, no interaction is possible among the criteria and the linear superposition holds. For a sub-additive measure, a redundant effect is modelled, while the contrary holds for a super-additive effect (synergic effect).

#### 3.2. The Choquet integral

It can also be written as:

This operator satisfies the following properties [Marichal, 1999a]:

*w*_{i} being the weight of the *i*th criterion.

For an intuitive explanation of the Choquet integral, see the example in Murofushi and Sugeno [1989].

#### 3.3. The Möbius transform and the dual values

Given a non-additive measure *m*, its dual values can be obtained from the following biunivocal Möbius transform [Grabish *et al*., 2003; Marichal, 1998]:

The inverse transform is given by:

To be the dual of a non-additive measure, the 2^{n} coefficients need to satisfy:8

It can be verified that the Choquet integral can be written in the dual space as:

Moreover, if *α*(*T*)>0, the coalition *T* is synergic (in average), if *α*(*T*)<0, it is redundant, if *α*(*T*) = 0, there is no interaction and the Choquet integral collapses into the WA operator [Marichal, 1998, 1999a, b].

From a computational point of view, given *n* criteria, a non-additive measure requires the assignment of 2^{n} coefficients, and this is very large as soon as *n* is greater than 5,6. In order to avoid this, the *k*-order models were introduced [Marichal, 1998], which assume interactions between subsets of cardinality less or equal to *k*, usually the second-order models are considered, that is, *k* = 2. Even though in many applications it can be reasonably assumed that there is no interactions between subsets with cardinality higher than 2, this hypothesis needs to be tested *a priori*.

#### 3.4. Andness and orness measures

Given a non-additive measure, it is possible to compute an *andness* measure together with its complementary *orness* measure. If the *andness* measure is close to 1, it means that the measure set tends to the MIN operator, which is to the logical conjunction of the criteria value, showing a conservative tendency of the Decision Maker (*pessimistic* behaviour). Conversely, if *orness* = 1 we obtain the MAX operator, the logical disjunction, a totally compensative operator, corresponding to an *optimistic* behaviour. The computation of the *orness* index in the dual space is given by:

Moreover:

Both indices can be easily computed given the dual values of the measure.

#### 3.5. Non-additive measures and the multi-linear operator

In the dual space, the Choquet integral computes, for each coalition, the minimum of the criteria values of the coalition. The MIN operator belongs to a wide class of operators, the *triangular norm* (T-norm), which satisfies a set of rationality properties and are widely used in the field of MCDA analysis, especially in the fuzzy logic applications [Klement *et al*., 2000]. Since the MIN is not compensative at all, some Authors proposed to substitute the MIN operator, in the dual space, with a smoother T-norm, [Kolesàrovà and Mordelovà, 2001; Klement *et al*., 2000; Despic and Simonovic, 2000; Fujimoto and Murofushi, 1997]. A natural choice can be the product of the values, that is a differentiable and partially compensative operator. We obtain the so-called *multi-linear* operator [Grabish *et al*., 2001]. In the dual space, substituting the MIN operator with the product, we obtain:

In the measure space the multi-linear operator has the following formulation [Marichal, 1999b]:

which represents a pseudo-Boolean function.

#### 3.6. Identification of the measures

As said above, one of the most critical point in the evaluation is the assignment of the numerical values of the non-additive measure. Many methods were presented in literature, but most of them are based either on quite complex optimization algorithms, or on data mining techniques. In this case study, we preferred a user-friendly approach, and adopted a method based on a suitable questionnaire [Despic and Simonovic, 2000]. Let us suppose that the DM(s) judgements are in the scale [0,100], with the usual meaning for the numerical values, i.e. 0 = WORST, 50 = MEDIUM, 100 = OPTIMAL and so on. For each criterion two particular extreme cases are enhanced, the OPTIMAL and the WORST ones, conventionally indicated with 1 and 0 respectively from now on. An *edge* is a (fictive) scenario formed by a combination of (only) WORST and OPTIMAL evaluation. Each edge is nothing else that a question that is asked to the DM(s), which will assign his/(their) evaluation in the scale [0,100]. The edges are the vertices of an hyper-polyhedron in the criteria space. It is sufficient to define the values only in all those vertex to obtain the values of the measure, and this is the minimum amount of information. This simplification causes a poor statistical robustness, since it corresponds to the minimum number of interpolating points in an *n*-dimensional space, but given the unavoidable uncertainty, which is implicit in every human decision process, this does not seem to be a serious obstacle, considering the information gain that should be obtained explicitly considering all the possible interactions among the criteria. The advantages with respect to the WA approach are evident.

Table I reports an instance of the questions that need to be formulated in the case of 3 criteria. Referring to the case study, we are considering the node in the Sustainability Tree, which evaluates the Sustainability starting from Intrinsic Sustainability, Economic-Financial Sustainability and Context. The fourth column reports the DM evaluation (only one DM is simulated here). For a better comprehension, the fourth row implements the question:

Table I. The valuation tableIntrinsic sustainability | Context sustainability | Economic and financial feasibility | Evaluation |
---|

*Sustainability* |

Worst | Worst | Worst | 0 |

Optimal | Worst | Worst | |

Worst | Optimal | Worst | |

Worst | Worst | Optimal | |

Optimal | Optimal | Worst | |

Optimal | Worst | Optimal | |

Worst | Optimal | Optimal | |

Optimal | Optimal | Optimal | 100 |

‘How would you score an hypothetical case where the Economic-Financial Sustainability is OPTIMAL, and the two other criteria, Intrinsic Sustainability and Context, are WORST?’

After having fulfilled all the answers, a simple algorithm computes the dual values and passes such parameters to a procedure that implements the computation of the multi-linear aggregator for a real case. Moreover, the *andness* and the *ornes*s degrees can be computed and the behavioural nature of the DM can be obtained.

Assume, for the previous example with three criteria represented in Table I, the ‘weight’ of the first criterion to be equal to 30, the second to 20 (the second and the third empty cells in the last column of the Table), while the ‘weight’ of the coalition formed by the two criteria together to be equal to 70 (the last empty cell in the last column). Then, a synergic effect can be observed, since the ‘weight’ of the coalition is greater than the sum of the weights of the single criteria.

Evaluation in intermediate points would increase the statistical robustness, but the numerical complexity of the algorithm would increase significantly either. We feel that the edges evaluation and the multi-linear operator are a good compromise choice between theoretical complexity and operative usefulness. Other solutions, see for instance [Fujimoto and Murofushi, 1997], are difficult to be implemented and require a strong computational effort. Moreover, the same approach can be used in the case of multi-person decision scenario, where many Experts or Decision Makers cooperate in the assignment of the ‘weights’ of the criteria coalitions, and a measure of consensus could be easily defined and computed [Kacprzyk, 1987; Kacprzyk and Fedrizzi, 1988; Kacprzyk *et al*., 1992].

### 4. EVALUATION OF SUSTAINABILITY OF RE-USE PROJECTS

- Top of page
- Abstract
- 1. EVOLUTION OF THE CONCEPT OF CULTURAL HERITAGE
- 2. THE SUSTAINABILITY OF THE RE-USE OF HISTORIC ARTEFACTS
- 3. MULTICRITERIA AGGREGATION
- 4. EVALUATION OF SUSTAINABILITY OF RE-USE PROJECTS
- 5. EVALUATION OF SUSTAINABILITY OF HYPOTHESIS FOR THE RE-USE OF THE HISTORIC VENETIAN ARSENAL
- 6. THE ASSESSMENT OF SUSTAINABILITY OF RE-USE PROJECTS
- 7. CONCLUSIONS
- Acknowledgements
- REFERENCES

In the previous paragraphs, we illustrated that integrated conservation is defined as the best possible compromise in dealing with conflicting objectives. Therefore, the operative phase of the study concentrated on the definition of indicators for the evaluation of sustainability of alternative re-use projects for historic artefacts.

The design of a hierarchy model for the evaluation of the sustainability was based on the definition of criteria synthesizing the main characteristics, which could influence the evaluation of sustainability. This initial phase was completed by consulting experts in urban re-qualification and the re-use of historic buildings. The resulting proposed indicators take into consideration the effects of the intervention on the artefact by using three main points of view: the impact on the historic building (defining future re-use—and relative standards—to be hosted in the historic building); the social impact; the economic and financial feasibility (Table II).

Table II. Hierarchic structure (simplified) of the evaluation model (* nodes) |

*Intrinsic sustainability*: or the respect of the materials and typology of the building. This criterion is the synthesis of three sub-criteria:

*Reversibility* or the opportunity to restore the building to the state it was in before the modifications carried out with the re-use project;

*Versatility* or the possibility to eventually modify the function of the building proposed by the re-use project without major works;

*Invasiveness* or the degree to which the project interferes with the materials the historic building is made of.

*Context sustainability*, which refers to the extent to which the reuse project enhances the social, economic and environmental context of the building and its contribution to the local identity. The re-use project must, where possible, rebuild a relationship between the building and its social and environmental setting. It is also hoped that the project will produce positive externalities on circulation and bring economic advantages to the territory.

*Economic and financial feasibility*, which evaluates the project according to economic and financial principles. The model implies that the objective of sustainable re-use also depends on the project's financial efficiency of the economic activity. Moreover, the risk concerning the investment must also be taken into account.

After evaluation model structuring, to each criteria, sub-criteria and attribute was given a weight, which defines its contribution towards sustainability. In order to calculate the weight of each single characteristic, a questionnaire was prepared applying the edge's method described in the previous paragraph.

The questionnaire had a page for each of the nodes of the hierarchical tree, so that to each leave belonging to the node would be given a weight. The questionnaire was compiled by 11 experts.

Table III presents the average score of the evaluation given by the experts, their standard deviation and variation coefficient for the ‘sustainability’ node and for the extreme scenarios.9

Table III. The scores attributed to the ‘sustainability’ nodeScenario | Sustainability | Average score | Standard deviation | Variation coefficient (%) |
---|

| Intrinsic sustainability | Context sustainability | Economic and financial feasibility | | | |
---|

1 | Worst | Worst | Worst | 0.0 | 0.0 | — |

2 | Optimal | Worst | Worst | 29.5 | 11.7 | 39.7 |

3 | Worst | Optimal | Worst | 24.7 | 19.3 | 78.1 |

4 | Worst | Worst | Optimal | 20.0 | 14.3 | 71.6 |

5 | Optimal | Optimal | Worst | 65.2 | 10.0 | 15.3 |

6 | Optimal | Worst | Optimal | 57.7 | 10.3 | 17.9 |

7 | Worst | Optimal | Optimal | 48.2 | 25.5 | 53.0 |

8 | Optimal | Optimal | Optimal | 100.0 | 0.0 | 0.0 |

The data presented puts into evidence that the experts place at the first level the intrinsic sustainability of the re-use project (Scen. 1), the coherence with the social context is placed in the second step (Scen. 2) and, finally, they consider the economic aspects (Scen. 3). Another interesting result is that the importance given to the intrinsic sustainability is quite stable across the experts' valuations (V.C. 40%), but they gave quite different evaluation scores for the indicators regarding the ‘context sustainability’ criterion and the economic–financial feasibility (V.C. 70–80%).

Analysing the scores given to scenarios (5, 6 and 7), where ‘optimal’ judgements are given contemporarily to two criteria, it emerges that a ‘optimal’ judgement given to the ‘intrinsic sustainability’ criterion is sufficient to realize a good (approx. 60) and stable (V.C. 10%) score. Furthermore, in the other case the overall score is low (48) and variable across the experts (V.C. 25%).

For instance, the sustainability index (*Is*) can be computed from the criteria scores reported in Table III using the Choquet aggregation formula (see Section 3.2).10

where *I*_{S} is sustainability; SI is intrinsic sustainability; *C* is context sustainability; FEF is economic and financial feasibility.

In the above equation the first three coefficients are simply the normalized values of rows 2, 3 and 4 of ‘Average score’ column in Table III. The other coefficients are the Möebius transform of the corresponding criteria coalition. For instance, the coefficient of the coalition (SI, *C*) is obtained applying the first formula of sub section 3.3 that is:

subsequently normalized dividing by 100.

Once the model has been calibrated with the value functions, the technician responsible for evaluating the sustainability of re-use projects expresses a judgement (0,100) for each parameters in which the various attributes of Table II have been disaggregated. This score is multiplied by the weight attributed to the parameter and by the weights assigned to the nodes higher up. In other words, giving a technical evaluation to each parameter associated with the project under examination, a comprehensive evaluation of the sustainability of the re-use project is realized.

The model is useful when there are several alternative projects to choose from, as it supplies a final sustainability score for each alternative and intermediate scores that refer to the criteria, sub-criteria and attributes. As described above, in order to assign weights to criteria, sub-criteria, attributes and parameters, the experts filled out a questionnaire and gave scores ranging from 0 to 100 to hypothetical scenarios.

The experts shared cultural knowledge in at least two fields: the *conservationists* were architects operating in the material restoration of historic buildings; the *designers* and *planners* were specialists in analysing and identifying the function that the historic building should be given and the economic evaluation of re-use.

It was thus useful to establish indices which would evaluate the attention toward conservation shown by each expert's judgement.

*Andness* and *orness* indices were used, where the index value may vary between 0 and 1 in both cases and takes on the following significance:

Total *andness*: the expert consulted considers that the sustainability of a project is guaranteed only if all the indicators are attributed the maximum score (*andness* index = 1; *orness* index = 0);

Total *orness*: the expert consulted considers that the sustainability of a project is guaranteed if one of the indicators is given the highest (*andness* index = 0; *orness* index = 1);

Mainly *andness*: the expert consulted considers that the sustainability of a project is guaranteed only if the majority of the indicators are attributed a high score (*andness* index>0.5; *orness* index <0.5);

Mainly *orness:* the expert consulted considers that the sustainability of a project is sufficiently guaranteed when one indicator rather than another receives a high score (*andness* index<0.5; *orness* index >0.5);

*Additive* measure: the expert consulted considers that the sustainability of a project depends on the sum of the scores assigned by the indicators, without there being any synergy between them (*andness* index = 0.5; *orness* index = 0.5).

Table IV shows average *Andness/Orness* indices for the 11 experts consulted. The majority of experts tended towards *Andness* behaviour in all the nodes examined which means that a project can be considered sustainable if at least two or more criteria are deemed ‘optimal’; thus it is not enough for the project to respect the historic building, but it must also be economically sustainable, and its reference context must be carefully considered. Furthermore, the *Andness* behaviour is higher for the ‘Intrinsic Sustainability’ criterion than in the other criteria.

Table IV. Indices of ‘*andness*’ and ‘*orness*’ for the most important criteria | Average | Std. Dev. | C.V. |
---|

*Sustainability* |

Orness | 0.409 | 0.087 | 0.214 |

Andness | 0.591 | 0.087 | 0.148 |

*Intrinsic sustainability* |

Orness | 0.493 | 0.079 | 0.16 |

Andness | 0.507 | 0.079 | 0.156 |

*Context* |

Orness | 0.501 | 0.05 | 0.099 |

Andness | 0.499 | 0.05 | 0.1 |

*Economic and financial feasibility* |

Orness | 0.48 | 0.071 | 0.149 |

Andness | 0.52 | 0.071 | 0.137 |

### 5. EVALUATION OF SUSTAINABILITY OF HYPOTHESIS FOR THE RE-USE OF THE HISTORIC VENETIAN ARSENAL

- Top of page
- Abstract
- 1. EVOLUTION OF THE CONCEPT OF CULTURAL HERITAGE
- 2. THE SUSTAINABILITY OF THE RE-USE OF HISTORIC ARTEFACTS
- 3. MULTICRITERIA AGGREGATION
- 4. EVALUATION OF SUSTAINABILITY OF RE-USE PROJECTS
- 5. EVALUATION OF SUSTAINABILITY OF HYPOTHESIS FOR THE RE-USE OF THE HISTORIC VENETIAN ARSENAL
- 6. THE ASSESSMENT OF SUSTAINABILITY OF RE-USE PROJECTS
- 7. CONCLUSIONS
- Acknowledgements
- REFERENCES

The model presented in the previous paragraphs has been used for the valuation of the sustainability of alternative re-use hypothesis of the ancient Arsenale of Venice.

The Venice Arsenale is owned by the Italian government and is currently used primarily by the Italian Navy. About 45 hectares in size, the Arsenale accounts for about 15% of the area of the city of Venice, and is located in the Castello district. Founded in 1104, in its heyday the Arsenale employed roughly 20 000 workers and was said to produce one ship a day.

The Arsenale started to decline after World War I, and continued to decline at an even faster rate after World War II, when its buildings were progressively abandoned. In 1983 the *Soprintendenza per i Beni Ambientali ed Architettonici* of Venice (local of office of the state authority for cultural heritage conservation) started a series of conservation works. At present, the Italian Navy continues to own and occupy a large portion of the Arsenale. Research activities, shipbuilding, museums and exhibitions occupy other areas, but many buildings and areas remain unutilized.

Out of the analysis of the political debate on alternative options for the re-use of the Arsenal, two basic alternative directions could be extrapolated. The first one is pointing to installing ‘poor’ functions in the ancient buildings without considering the historic significance of the area, but well compatible with the historic building structures. The functions to be introduced are small artisans activities (carpenters, electricians, masons etc.) mostly already working within the historic centre but often under menace of expulsion because of pressings from the real estate market. The second option points to the introduction of ‘new’ uses somehow connected to the Arsenal's historic marine function, like a touristic marina. On the basis of these basic assumptions, two hypothetic projects or scenarios have been created in order to evaluate their sustainability.

#### 5.1. 1st Scenario: area for artisans

In the first scenario it is assumed to use the buildings of the Arsenale for craftsmen's activities actually dispersed in the historic centre. The surfaces of water of the main dock and some of the buildings will be used for laying up small boats owned by Venetian residents.

It is presumed that the whole surface and all buildings, except those actually occupied by the Navy, will be used by artisan's activities. The re-conversion will take place after a restoration programme managed by the municipality, which will adapt the buildings to the requirements of craftsmanship and small manufacturing activities. The industries that are going to settle within the restored buildings will pay a rent ruled by medium- to long-term contract (around 20 years). The surface of the big dock (Darsena Grande) will be used for mooring of boats owned by Venetian residents. A limited number of buildings, including the covered docks, will be used for mooring and laying up of boats on high rise racks.

#### 5.2. 2nd Scenario: marina

The second scenario refers to a proposal frequently presented in the past to use the historic Arsenale as a touristic marina for permanent and temporary mooring. The activities to be introduced regard, beyond the berths themselves (approx. 220 places), supplementary facilities comprising high quality shipyards, boats repair and laying up services, shops and services necessary for tourism, as retail stores for nautical equipment.

In this proposal the area's original shipyard vocation is taken up, expecting the nautical tourism to contribute to a revival of the traditions of this place in terms of boatbuilding. The berths of the main dock will be partly assigned on a permanent basis, 25% will be reserved for temporary mooring.

The historic buildings will house the facilities connected to the port such as marine shops, craftsmen activities and boatbuilding as well as a shipyard for the production of leisure time boats. A supermarket will be located in a position easy to access from the surrounding residential areas as well.

The open spaces, transformed in quays, are used as slipways for the marine activities and shipyards.

Some buildings on the southern front of the main dock will be transformed in to reception area with restaurants and bars, a yacht club, and rooms for small events, sailing schools etc. as well as services offering assistance for guests.

Introducing productive activities into the historic buildings does not represent a particular problem from the conservation point of view. Some more problems may be represented by the introduction of commercial facilities and supermarkets, which might ask for divisions of the inner spaces, with consequent modification of the typologies of the historic buildings.

### 7. CONCLUSIONS

- Top of page
- Abstract
- 1. EVOLUTION OF THE CONCEPT OF CULTURAL HERITAGE
- 2. THE SUSTAINABILITY OF THE RE-USE OF HISTORIC ARTEFACTS
- 3. MULTICRITERIA AGGREGATION
- 4. EVALUATION OF SUSTAINABILITY OF RE-USE PROJECTS
- 5. EVALUATION OF SUSTAINABILITY OF HYPOTHESIS FOR THE RE-USE OF THE HISTORIC VENETIAN ARSENAL
- 6. THE ASSESSMENT OF SUSTAINABILITY OF RE-USE PROJECTS
- 7. CONCLUSIONS
- Acknowledgements
- REFERENCES

The aim of the paper has been to present a procedure for the evaluation of the sustainability of projects for the economic re-use of historical buildings in Venice. A multiple criteria model for the analysis of alternative projects for re-use and to support the choice was set up. The model adopts a hierarchical approach that identifies the relevant indicators for the appraisal of sustainability, and groups them into three criteria: intrinsic sustainability, context sustainabilitya and economic-financial feasibility. The aggregation operator at each node of the hierarchical tree of the model computes a global evaluation based on non-additive measures and the multi-linear aggregation function. The measure values are implicitly obtained from a panel of experts who filled a questionnaire on hypothetical scenarios, allowing for the calibration of the value function with which to analyse the sustainability. The preference structure obtained permits the analysis of the conjunctive–disjunctive (andness–orness) behaviour of the experts.

Starting from the opinions expressed, indicators were then drawn up to estimate the level of conservativeness of the expert evaluations.

Operationally, the evaluation model was tested on two reuse hypothesis of the Arsenal in Venice. The evaluation model seems able to provide interesting results on the sustainability of the projects for re-use, correctly considering the environmental, social and economic components of the work and highlighting the strengths and weaknesses of the two type of re-use. Such analysis can be used in various ways.

Primarily, it can provide a useful support for the identification of the critical point, at the preliminary stage, of projects capable of combining conservation and economic improvement. Second, it can be a support for the selection of projects to be financed in that it allows the trade-off between economic use and conservation to be appraised and thus, implicitly, the cost of the conservation. Finally, it can provide a means of reading the projects for re-use, a kind of checklist of variables to be considered in the evaluation of the proposals.