The use of a multi-attribute tool for evaluating accessibility in buildings: the AHP approach
The Authors
S. Wu, School of Construction and Property Management, University of Salford, Salford, UK
A. Lee, School of Construction and Property Management, University of Salford, Salford, UK
J.H.M. Tah, School of Construction and Property Management, University of Salford, Salford, UK
G. Aouad, School of Construction and Property Management, University of Salford, Salford, UK
Abstract
Purpose – The purpose of this article is to develop a quantitative building accessibility assessment model for the construction industry.
Design/methodology/approach – The building accessibility assessment criteria are incorporated in a hierarchy structure based on the relevant building regulations and British standards. The analytic hierarchy process (AHP) is employed to determine the priority of the accessibility criteria. A review of the application of AHP is included in the paper. Finally, a case scenario is used to illustrate the method.
Findings – This paper provides a methodology to prioritize the building accessibility criteria and to indicate how well a building design meets accessibility requirements quantitatively.
Practical limitations/implications – A model is advocated for use by accessibility consultants and building designers to establish a quantitative assessment for building accessibility. It can also be used in the development of accessibility assessment software.
Originality/value – This paper presents a novel quantitative building accessibility assessment model.
Article Type:
Research paper
Keyword(s):
Analytical hierarchy process; Buildings.
Journal:
Facilities
Volume:
25
Number:
9/10
Year:
2007
pp:
375-389
Copyright ©
Emerald Group Publishing Limited
ISSN:
0263-2772
Introduction
Current UK legislation regarding accessibility is governed by Building Regulations and the Disability Discrimination Act (1995), which highlights the need for inclusive environments for all people irrespective of any disability, including people with physical, sensory and cognitive impairments. The essential requirements of DDA were introduced in the UK in September 2004. Since this date all service providers have been required to make “reasonable adjustment” to the physical features of their premises to overcome physical barriers to access' (Sawyer and Bright, 2004). The new edition of part M of the building regulations came into force in May 2004. It was revised to take account of the guidance given in BS8300:2001, “Design of buildings and their approaches to meet the needs of disable people – code of practice” (ODPM, 2004).
At present, the accessibility of an existing building or new development is assessed through access audit and access appraisals; this is in addition to the access officer from the local planning office involved in the development of public facilities during the planning stage:
- Access audit. Establishes how well a particular building or environment performs in terms of access and ease of use by a wide range of potential users. The audit report should recommend access improvements, prioritise action and indicate where improvements can be made (Sawyer and Bright, 2004).
- Access appraisal. Audit of the proposals for a new development, refurbishment or alteration. This involves making a detailed assessment of the proposed level of accessibility in a building using drawings, specifications and consultation with architect or designer (Sawyer and Bright, 2004).
The process of the both access appraisal and auditing involves a thorough site inspection, an assessment of the management and use of the building and preparation of a report that identifies accessible user-friendly feature as well as access problems. However, the process of this assessment is quite complex and often includes a very long checklist with over a hundred design criteria has to be completed. Furthermore, accessibility design criteria is not limited to the quantifiable requirement, such as physical dimensions of a door or space, it also includes a large number of subjective requirements, such as “edges of glazed door should clearly visible, door furniture should be distinguishable.” The assessment of such subjective requirements is heavily depended on the experience of the assessor. As part of the access appraisal and auditing, a prioritised recommendation is included. It is extremely difficult to set appropriate priority under different scenarios when a large number of criteria are involved. It is also difficult to provide a clear indicator for the client or designer about the overall accessibility level of the building/environment at the end of the assessment process. This research aims to develop a quantitative indicator for the overall accessibility level of a building/environment. A number of multi-attribute methodologies have been investigated, such as goal programming, linear programming and analytic hierarchy process (AHP). The AHP was chosen to meet the accessibility assessment requirement to support the highly structured access design criteria and the capability to produce appropriate priority for each criterion.
A review of analytic hierarchy process (AHP)
The AHP method was developed by Thomas Saaty more than two decades ago for elucidating and resolving unstructured problems in the economic, social and management sciences. As Saaty (1980) stated:
To be realistic our models must include and measure all important tangible and intangible, quantitatively measurable, and qualitative factors.
Methodologically, it combines the basics of qualitative and quantitative research to solve decision problems by justifying the decision-making process. It is described by Partovi (1994) as:
A decision-aiding tool for dealing with complex, unstructured and multi-attribute decision.
Muralidhar et al. (1990) support the belief that AHP particularly caters for decision making with multi-criteria. Apart from this, the high precision of relative priorities in the calculations enhances the effectiveness of this technique.The applications of AHP have been applied in industry to solve commercial decision problems and address empirical research issues (Easley et al., 2000). Decisions today are more complicated and difficult to make due to the greater number of impacts on them (e.g. larger set of factors or criteria) and severe consequences resulting from poor decisions (De Boer et al., 2001). The AHP method is expected to circumvent other basic linear weighting methods to deal with imprecision for complex problems (De Boer et al., 1998).
AHP has been extensively applied in different areas including marketing, finance, education, public policy, economics, medicine, sports, amongst others (Saaty et al., 1987; Saaty and Mu, 1997; Saaty and Nezhad, 1981; Saaty and Rush, 1987) because of the ease of its use. New applications have been found in the fields of information and management (Byun, 2001; Forgionne and Kohli, 2001; Lai et al., 1999; Yang and Huang, 2000). AHP has been applied more recently in construction research (Li et al., 2000). Apart from its value in solving a decision problem (e.g. selection of a contractor), its usage has been extended to prioritise elements in a survey environment. For example, Tan and Lu (1993) applied AHP when prioritising the criteria and factors affecting the quality of construction engineering design projects. Its popularity comes from three main advantages as described by Saaty (1980):
- It helps to decompose a complex and unstructured real world multiple criteria decision making problem (or research problem) into a set of elements in terms of variables organised in a multilevel hierarchical form that also determines the overall priorities by quantifying information providers' subjective judgements.
- It employs a pairwise comparison process by comparing two objects at a time to formulate a judgement as to their relative weights. As this method exhaustively compares one element with others, it can generate more useful information available to validate the results.
- It measures the consistency level of each judgement matrix. Some researchers refer to the consistency measure as the consistency test (Cheng and Li, 2001; Leung and Cao, 2001). Specially, with adequate measurements, the AHP is more accurate (with fewer experimental errors) in achieving a higher degree of consistency.
Despite the above advantages, the effectiveness of AHP as a decision tool has been the subject of much debate. For example, as it uses a scale value from 1 to 9 for pairwise comparisons. Decision makers or respondents may need time to compare all of the paired elements, especially when the problem of study consists of many levels, where the elements of each level further divide into many sub elements. It is also argued that indication of the consistency level is not necessary when the information providers are clear about what they want to rank.
A study by Easley et al. (2000) supported that other pairwise comparison methods (except the ordinal paired comparison method) may be as accurate as AHP for making group decisions, but was either less difficult to use or allowed for no consistency measure of responding to matrixes. They suggested the paired hierarchical (ranking) method, which relaxes the imposed 1-9 scale constraint, and the folded normal AHP (FNAHP) (MacKay et al., 1996), which relaxes the reciprocal comparison required for computing the consistency ratio. Nevertheless, there is a basic assumption of these two methods that the consistency in responses must still be maintained, but it is pre-assured rather than post measured (Easley et al., 2000). Moreover, owing to the prevalence of using a simple scale to rate a group of multiple criteria in questionnaire-based research, the answers may be misconstrued without paired comparison, resulting in greater value of the consistency measure. Further, a study by Cheng and Li (2001) concluded that the consistency measure is a critical component of AHP, and it makes AHP more reliable and useful as decision-making tool.
The accessibility of a building is determined by a large number of criteria, such as the size of the door, size of the corridor, attitude of staff, etc. Current assessment methods have to check every single criterion, but it does not highlight the relative importance of each criterion, although an experienced access consultant is aware of the priorities of the requirements.
Furthermore, the final assessment result is usually presented as a report, and does not provide a direct and quantifiable indication of the overall accessibility level of the building. It is difficult for the owner of an existing building to prioritise their resource to make reasonable adjustment, and it is also difficult for designers to understand the accessibility level of their new design. Therefore, a solution to prioritise the accessibility criteria and be able to provide quantitative indicator is needed. In this paper, AHP is employed to tackle this problem.
AHP methodology
AHP is a hierarchical representation of a system. A hierarchy is an abstraction of the structure of the system as a result of the decomposition of the complexity of the system into different levels, which represent functional interactions of its elements and their impacts on the entire system (Saaty, 1980). In order to develop the accessibility indicator with the AHP, the following five steps are taken:
- Define the building accessibility indicator.
- Construct a hierarchy of criteria affecting the accessibility indicator.
- Employ a pair-wise comparison method for the criteria.
- Compute the consistency level to drop out the inconsistent responses.
- Compute relative weights of each criterion.
Step 1: define the accessibility indicator
The building accessibility indicator is a methodology to establish relative importance of the accessibility design criteria and a quantitative measure for the overall accessibility level of a building.
Step 2: construct the hierarchy of the accessibility design criteria
The British Standards (BS 8300:2001), the revised Part M of the building regulation and the provisions of DDA (1995) have been reviewed to establish the detailed design criteria.
The BS 8300:2001 design of buildings and their approaches to meet the needs of disabled people – code of practice gives detailed guidance on the design of domestic and non-domestic buildings. It draws on research commissioned by the Department of Environment, Transport and the Regions in 1997 and 2001 and it is the most comprehensive standard to date covering the needs of people with disabilities.
The new edition of Part M of the building regulations refers to new and existing buildings being accessible and usable by people and takes account of the guidance given in BS 8300:2001.
Based on these requirements, the accessibility design criteria are structured into the following four levels to form an accessibility assessment decision hierarchy (Figure 1).
Level I: it is the objective or the overall goal of the accessibility assessment, which is to ensure the accessibility of a building and provide an indication of the overall accessibility level of a building. The name given to level I of the hierarchy is the “overall accessibility level”.
Level II: the second level represents the scope of the accessibility assessment. The access audit or appraisal usually covers two main areas, namely:
- Physical feature.
- Access management issues.
Physical feature includes (Sawyer and Bright, 2004):
- any feature arising from the design or construction of a building on the premises occupied by the service provider;
- any feature on those premises or any approach to, exit from or access to such a building; and
- any fixtures fittings, furnishings, furniture, equipment or materials in or on such premises.
Access management issues are dealt with the access to information, staff attitudes towards disability and accessibility issues, management procedure, maintenance and use of the building.
Level III: this level breaks level II into further detailed elements, which is related to individual areas and specific functions of a facility. Table I shows the detailed items identified as level III items (Table I).
Level IV: this level details the key building components affecting accessibility assessment, which are represented at the fourth level of the hierarchy that contains detailed access criteria of each building components. In total, 42 components are included in the assessment criteria hierarchy (see Figure 1). It will be difficult to make a pair-wise comparison with many criteria at multiple levels (Tam and Tummala, 2001). Therefore, further detailed checklists are not included in the criteria hierarchy.
Step 3: employ pair-wise comparison
Once the assessment criteria hierarchy has been constructed, the next step is to determine the priorities of elements at each level (“element” here means every member of the hierarchy). However, before starting the AHP process, it is important to consider how the building is used, managed and operated. A building often contains many areas where different functions may affect the access requirement. The entire building or parts of the building can be classified according to its use. It is also helpful at this stage to understand how it can be improved to suit the needs of the users. There are four use classifications described in Table II. Furthermore, an analysis of the building using this classification can help to give a more accurate picture of the specific access requirements of each individual area and appropriate judgement can be made in the following AHP comparison process.
To begin the AHP process, a set of comparison matrices of all elements in a level of the accessibility criteria hierarchy with respect to an element of the immediately higher level are constructed so as to prioritise and convert individual comparative judgements into ratio scale measurements. The preferences are quantified by using a nine-point scale. The meaning of each scale measurement is explained in Table I. The pairwise comparisons are given in terms of how much element A is more important than element B (Table III). As the AHP approach is a subjective methodology, information and the priority weights of elements may be obtained from the decision-maker of the company using direct questioning or a questionnaire method. In the case example, the data is acquired through discussion within an accessibility expert, disable users and the research team. Tables IV to VII are the result of the pair comparison matrix for both “physical feature” and “management issues”.
Step 4: computing the consistency level
The pairwise comparisons generate a matrix of relative rankings for each level of the hierarchy. Table V shows the relative ranking for each element in the physical feature, and Table VII shows the relative ranking for each element in management issues.
The number of matrices depends upon the number of elements at each level. The order of the matrix at each level depends on the number of elements at the lower level that it links to. After all matrices are developed and all pairwise comparisons are obtained, eigenvectors or the relative weights (the degree of relative importance amongst the elements), global weights, and the maximum eigenvalue (λmax) for each matrix are then calculated in spreadsheet.
The λmax value is an important validating parameter in AHP. It is used as a reference index to screen information by calculating the consistency ratio CR (Saaty, 2000) of the estimated vector in order to validate whether the pair-wise comparison matrix provides a completely consistent evaluation. The consistency ratio is calculated as per the following steps:
(1) Calculate the eigenvector or the relative weights and λmax for each matrix of order n.
(2) Compute the consistency index for each matrix of order n by the formulae:
(3) The consistency ratio is then calculated using the formulae:
where RI is a known random consistency index obtained from a large number of simulation runs and varies depending upon the order of matrix. Table VIII shows the value of the random consistency index (RI) for matrices of order 1 to 10 obtained by approximating random indices using a sample size of 500 (Saaty, 2000).
It has been a difficult process to maintain consistency in judgement during these exercises when the size of paired comparison matrix increased. The consistency ratio of the matrix of physical features is 0.046 and the size of the matrix is 7. This comparison matrix had to be completed twice in order to achieve a consistency judgement. Therefore, it is essential to group the related criteria into several groups and each group contains no more than ten sub criteria.
Step 5: computing relative weights of each criterion
Saaty (1996) points out that “if there are more than two levels, the various priority vectors can be combined into priority matrices, which yield one final priority vector for the bottom level”. Local priority is the priority relative to its parent. Table IX shows the priority of each criterion in the final selection of a contractor. Global priority, also called final priority, is the priority relative to the goal.
As mentioned earlier, detailed checklists for each of the building components are not included in the criteria hierarchy to reduce the number of pairwise judgments. In order to provide a quantitative measurement at this bottom level, a five-point rating (Liberatore et al., 1992; Liberatore, 1987) is introduced into this model. The alternative approach is to use the five-rating score of outstanding (O=1), good (G=2), average (A=3), fair (F=4) and poor (P=5) to rate each of the building components. Also, this helps to decrease unexpected bias that might occur in the process of decision-making when there are a large number of sub-factors to be compared.
In this paper, a demonstration exercise is conducted based on a building with “complete freedom of movement” use classification. Table IX shows the global weight of all level 4 criteria. The entrance doors (0.0501), WCs (0.0409) and reception area (0.0438) are the three most important criteria of physical features for our scenario building. These results are acknowledged by our accessibility experts in the research team. The criteria of “access to information” and “management policy and practices” under access management also have high global weight, which are 0.1859 and 0.0878, that is because these criteria are not detailed further in level 4 criteria. The exercise also demonstrated how to use the five-rating method to calculate the overall accessibility indicator for the whole building. The final indicator is 2.96, which means that this building meets about 60 per cent of the accessibility requirements.
Conclusion
New legislation and building regulations have made building accessibility even more important than before. A quantitative measure of building accessibility is needed to help clients and designers to understand accessibility issues and manage the services to comply with the new requirements. This research proposed “the accessibility indicator” as a methodology using the AHP method to help determine the relative importance of all the accessibility criteria and produce quantitative measure as a result. It is not an absolute indicator for building accessibility, but aims to represent how well a building meets the specified accessibility requirements based on the legislation, building regulation and user requirements.
A total of 54 criteria were collected from the building regulations, British Standards, and various accessibility literatures. They have been structured into a hierarchy which AHP method can be performed on. However, the criteria was still too large and was difficult to maintain consistent judgements for the paired comparison exercise. The next step of the research is to establish a large set of data to provide basic indicators for different types of buildings under different scenarios and users can use these indicators as a base to customise their criteria to suit their specific requirements. Also computer tool needs to be developed to manage the data and assist the comparison exercise.
Equation 1
Equation 2
Figure 1Accessibility criteria hierarchy for physical features
Table ILevel II and III accessibility criteria hierarchy
Table IIUse classification
Table IIIScale of measurement in pairwise comparison
Table IVPaired comparison for physical feature
Table VNormalised matrix for physical feature
Table VIPaired comparison for management issues
Table VIINormalised matrix for management issues
Table VIIIAverage random index (RI) based on matrix size
Table IXLocal weight and global weight
References
Byun, D. (2001), "The AHP approach for selecting an automobile purchase model", Information and Management, Vol. 38 No.5, pp.289-97.
Cheng, E.W.L., Li, H. (2001), "Information priority setting for better resource allocation using analytic hierarchy process (AHP)", Information Management & Computer Security, Vol. 9 No.2, pp.61-70.
De Boer, L., Labro, E., Morlacchi, P. (2001), "A review of methods supporting supplier selection", European Journal of Purchasing and Supply Management, Vol. 7 No.2, pp.75-89.
De Boer, L., Van der Wegen, L., Telgen, J. (1998), "Outranking methods in support of supplier selection", European Journal of Purchasing and Supply Management, Vol. 4 No.2/3, pp.109-18.
Disability Discrimination Act (1995), available at: www.hmso.gov.uk/acts/acts1995/ (accessed January 2004), .
Easley, R.F., Valacich, J.S., Venkataramanan, M.A. (2000), "Capturing group preferences in a multicriteria decision", European Journal of Operational Research, Vol. 125 pp.73-83.
Forgionne, G.A., Kohli, R. (2001), "A multiple criteria assessment of decision technology system journal quality", Information and Management, Vol. 38 No.7, pp.421-35.
Lai, V.S., Trueblood, R.P., Wong, B.K. (1999), "Software selection: a case study of the application of the analytical hierarchical process to the selection of a multimedia authoring system", Information and Management, Vol. 36 No.4, pp.221-32.
Leung, L.C., Cao, D. (2001), "On the efficacy of modeling multi-attribute decision problems using AHP and Sinarchy", European Journal of Operational Research, Vol. 132 pp.39-49.
Li, H., Cheng, E.W.L., Love, P.E.D. (2000), "Partnering research in construction", Engineering, Construction and Architectural Management, Vol. 7 No.1, pp.76-92.
Liberatore, M.J. (1987), "An extension of the analytical hierarchy process for industrial R&D project selection and resource allocation", IEEE Transactions on Engineering Management, Vol. 34 No.1, pp.12-18.
Liberatore, M.J., Nydick, R.L., Sanchez, P.M. (1992), "The evaluation of research papers (or how to get an academic committee to agree on something)", Interfaces, Vol. 22 No.2, pp.92-100.
MacKay, D.B., Bowen, W.M., Zinnes, J.L. (1996), "A Thurstonian view of the analytic hierarchy process", European Journal of Operational Research, Vol. 89 pp.427-44.
Muralidhar, K., Santhnam, R., Wilson, R.L. (1990), "Using the analytic hierarchy process for information system project selection", Information and Management, No.February, pp.87-95.
ODPM (2004), "Access to and the use of buildings – Part M of The Building Regulation 2000", The Office of the Deputy Prime Minister, London, .
Partovi, F.Y. (1994), "Determining what to benchmark: an analytic hierarchy process approach", International Journal of Operations & Production Management, Vol. 14 No.6, pp.25.
Saaty, T.L. (1980), The Analytic Hierarchy Process, McGraw-Hill, New York, NY, .
Saaty, T.L. (1996), Multicriteria Decision Making, RWA Publications, Pittsburgh, PA, .
Saaty, T.L. (2000), Fundamentals of Decision Making and Priority Theory, 2nd ed., RWS Publications, Pittsburgh, PA, .
Saaty, T.L., Mu, E. (1997), "The Peruvian hostage crisis of 1996-1997: what should the government do?", Socio-economic Planning Sciences, Vol. 31 No.3, pp.165-72.
Saaty, T.L., Nezhad, H.G. (1981), "Oil prices: 1985 and 1990", Energy Systems and Policy, Vol. 5 pp.303-18.
Saaty, T.L., Rush, M. (1987), "A new macroeconomic forecasting and policy evaluation method using the analytic hierarchy process", Mathematical Modelling, Vol. 9 pp.219-31.
Saaty, T.L., Blair, A., Nachtmann, R., Olson, J. (1987), "Forecasting foreign exchange rates: an expert judgement approach", Socio-economic Planning Sciences, Vol. 21 No.6, pp.363-9.
Sawyer, A., Bright, K. (2004), The Access Manual, Blackwell Publishing, London, .
Tam, M.C.Y., Tummala, V.M.R. (2001), "An application of the AHP in vendor selection of a telecommunications system", Omega, Vol. 29 No.2, pp.171-82.
Tan, R.R., Lu, Y. (1993), "On the quality of construction engineering design projects: criteria and impacting factors", International Journal of Quality & Reliability Management, Vol. 12 No.5, pp.18-37.
Yang, C., Huang, J. (2000), "A decision model for IS outsourcing", International Journal of Information Management, Vol. 20 No.3, pp.225-39.
Further Reading
Atthirawong, W., MacCarthy, B. (2001), "Identification of the location pattern of manufacturing plants in Thailand", proceedings of the 6th International Manufacturing Research Symposium, Cambridge, UK, 9-11 September, pp.1-13.
Badri, M.A., Davis, D.L., Davis, D. (1995), "Decision support models for the location of firms in industrial sites", International Journal of Operations & Production Management, Vol. 15 No.1, pp.50-62.
Blalock, H.M. Jr (1988), Social Statistics, 2nd ed., McGraw-Hill Book Co., Singapore, .
Canada, J.R., Sullivan, W.G. (1989), Economic and Multi-attribute Evaluation of Advanced Manufacturing Systems, Prentice-Hall, Englewood Cliffs, NJ, .
Centre for Accessible Environments (1999), Deigning for Accessibility: An Essential Guide for Public Buildings, Centre for Accessible Environments, London, .
Cheng, E.W.L., Li, H., Love, P.E.D. (2000), "Establishment of critical success factors in construction partnering", Journal of Management in Engineering, ASCE, Vol. 16 No.2, pp.84-92.
Cheng, E.W.L., Li, H., Love, P.E.D., Irani, Z. (1999), "The development of a conceptual framework for structuring information in construction projects", 9th Annual BIT Conference: Business Information Technology Management Generative Futures, CD Proceeding, The Manchester Metropolitan University, 3 and 4 November 1999, .
Cheng, E.W.L., Li, H., Love, P.E.D., Irani, Z. (2002a), "Development of a generic conceptual framework for a total information networking system", International Journal of Services, Technology and Management, Vol. 3 No.1, pp.125-38.
Cheng, E.W.L., Li, H., Shen, L.Y., Fong, P.S.W. (2002b), "Ranking of construction information for various functions of the general contractors: a preliminary study", Journal of Construction Research, Vol. 3 No.1, pp.181-91.
Crowe, T.J., Noble, J.S., Machimada, J.S. (1998), "Muti-attribute analysis of ISO 9000 registration using AHP", International Journal of Quality & Reliability Management, Vol. 15 No.2, pp.205-22.
Hafeez, K., Zhang, Y., Malak, N. (2002), "Determining key capabilities of a firm using analytical hierarchy process", International Journal of Production Economics, Vol. 76 No.1, pp.39-51.
Hoffman, J.J., Schniederjans, M.J. (1994), "A two-stage model for structuring global facility site selection decisions: the case of the brewing industry", International Journal of Operations & Productions Management, Vol. 14 No.4, pp.79-96.
Jungthirapanich, C., Benjamin, C.O. (1995), "A knowledge-based decision support system for locating a manufacturing facility", IEE Transactions, Vol. 27 pp.789-99.
Krajewski, L.J., Ritzman, L.P. (2002), Operations Management: Strategy and Analysis, 6th ed., Addison-Wesley, Boston, MA, .
Lam, K., Zhao, X. (1998), "An application of quality function deployment to improve the quality of teaching", International Journal of Quality & Reliability Management, Vol. 15 No.4, pp.389-413.
McCaffer, R., Baldwin, A.N. (1984), Estimating and Tendering for Civil Engineering, Granada, London, .
MacCarthy, B., Atthirawong, W. (2001), "Critical factors in international location decision: a Delphi study", Proceedings of 12th Annual Meeting of the Production and Operations Management, Florida, USA, 30 March-2 April, 2001, .
Ndekugri, I.E., McCaffer, R. (1988), "Management information flow in construction companies", Construction Management and Economics, Vol. 6 pp.273-94.
Platts, K.W. (1993), "A process approach to researching manufacturing strategy", International Journal of Operations & Production Management, Vol. 13 No.8, pp.4-17.
Platts, K.W., Mills, J.F., Bourne, M.C.S., Neely, A.D., Richards, A.H., Gregory, M.J. (1998), "Testing manufacturing strategy formulation processes", International Journal of Production Economics, Vol. 56/57 pp.517-23.
Saaty, T.L. (1994a), Decision Making in Economic, Political, Social, and Technological Environments with the Analytic Hierarchy Process, RWA Publications, Pittsburgh, PA, .
Saaty, T.L. (1994b), "Highlights and critical points in the theory and application of the analytic hierarchy process", European Journal of Operational Research, Vol. 74 pp.426-47.
Saaty, T.L. (1994c), "How to make a decision: the analytic hierarchy process", Interfaces, Vol. 24 No.6, pp.19-43.
Saaty, T.L. (1995), Decision Making for Leaders: The Analytical Hierarchy Process for Decision in a Complex World, RWA Publications, Pittsburgh, PA, .
Saaty, T.L. (1996), Decision Making with Dependence and Feedback: The Analytic Network Process, RWA Publications, Pittsburgh, PA, .
Slack, N., Lewis, M. (2002), Operations Strategy, Prentice-Hall, Harlow, .
Yang, J., Lee, H. (1997), "An AHP decision model for facility location selection", Facilities, Vol. 15 No.9/10, pp.241-54.
Zahedi, F. (1986), "The analytic hierarchy process: a survey of the method and its applications", Interfaces, Vol. 16 No.4, pp.96-108.
Corresponding author
S. Wu can be contacted at: s.wu@salford.ac.uk