Application of an Analytical Hierarchy Process to Prioritize the Factors Affecting ERP Implementation

Application of an Analytical Hierarchy Process to Prioritize the Factors Affecting ERP Implementation M.N. Vijaya kumar Lecturer, Dept. of Industrial Engg & Management, R.V. College of Engineering, Bangalore, Dr. A.V. Suresh Prof. & Head, Dept. of Industrial Engg & Management, R.V. College of Engineering, Bangalore, Dr. K.N. Subramanaya Asst. Professor, Dept. of Industrial Engg & Management, R.V. College of Engineering, Bangalore, ABSTRACT An enterprise resource planning (ERP) system is an integrated software solution, The very fact that about 60-70% of all ERP implementations in organizations around the world end up in a failure or meet their end prematurely calls for a serious look into the factors leading to such results of implementing a powerful integrating tool such as an ERP system, Literature review also shows that the AHP is the preeminent slant among the various methodologies applied to ERP projects in the past for prioritizing the attributes. Hence, in this paper we have applied the analytical hierarchy process (AHP) to Factors affecting ERP implementation. The model is based on the analysis of quality issues that affect a implementation process. To understand some criticalities of the model, a survey based on Analytic Hierarchy Process was carried out. Two sets of expert were selected i.e. End user and Third party vendor. On the basis of the results obtained, the epitome of findings is discussed. Index Terms ERP Implementation, AHP Modeling. I. INTRODUCTION Enterprise Resource Planning (ERP) is an enterprise-wide resources, information, and activities needed to complete business processes such as order fulfillment or billing. ERP systems can cover a wide range of functions and integrate them into one unified database. For instance, functions such as Sales and Distribution, Financial accounting and costing, Material Management, Human Resources, Production Planning, were all once stand alone software applications. Almost all organizations are turning to some sort of enterprise to some of enterprise resource planning package as a solution to their information management problems.erp package, if chosen correctly, implemented judiciously and used efficiently,will raise the productivity and profits of companies dramatically. But many companies fail in this because of wrong product, incompetent and haphazard implementation and inefficient usage. This study is carried out in a small and medium scale industries (SMEs) which is into agricultural and biotech research and product manufacturing to help and contribute largely to the farmer community in increasing their productivity year after year keeping the fertility of the soil, the organization under study had already gone for an ERP implementation and failed in go live situation. Company was planning for re implementation and elimination of root causes for the failure of the earlier implementation. Thomas L. Saaty (Saaty, 1980) evolved the analytical hierarchy process (AHP). It is a methodology for multi-criteria analysis and decision-making which can enable decision makers to represent the interaction of multiple factors in complex situations. The process requires the decision makers to develop a hierarchical structure for the factors which are explicit in the given problem and to provide judgments about the relative of each of these factors to specify a preference for each decision alternative with respect to each factor. It provides a prioritized ranking order indicating the overall preference for each of the decision alternatives. AHP uses a hierarchy to structure a decision problem, which deconstructs the problem into its component elements, groups the elements into homogeneous sets and arranges them hierarchically. Based on the hierarchical model, the AHP provides a method to assign numerical values to subjective judgments on the relative of each element and then to synthesize the judgments to determine which elements have the highest priority. The general approach of AHP is to decompose the total problem into smaller sub-problems in such a way that each sub-problem can be analyzed and appropriately handled with practical perspectives in terms of data and information. The objective of decomposition of the total problem into several levels is to enable pairwise comparisons of all the elements on a given level with respect to the related elements in the level just above. The solution process consists of three stages: 1. Determination of the relative of the attributes 2. Determination of the relative of each of the alternatives with respect to each attribute 3. Overall priority weight determination of each of these alternatives Following are the merits of the AHP: It has the ability to mix qualitative and quantitative criteria in the same decision framework It has the ability to integrate with techniques like goal programming It is possible to incorporate risk factors in the AHP It is very easy to incorporate sensitivity analysis in the AHP II. LITERATURE REVIEW The Implementation of new technologies and manufacturing philosophies in industrial sector with good success rates is crucial in a nation s economic growth and prosperity. ERP is one such system for which a lot of resistance is offered in organizations for implementation due to higher investments and more failures associated with it. The study of ERP implementation issues is necessary to encourage and persuade small and medium scale industries to go for ERP implementation as ERP is vital in their future growth. Many studies in the past on ERP have focused on implementation issues in SMEs. The critical success factors (CSF s) for a successful implementation of an ERP project has been evaluated and the role and effort on the part of all the stake holders involved are highlighted in these studies. These studies have been carried out in 1

different countries with different economic policies and financial growth rates. In case of small and medium scale industries (SMI) of India the study revealed that for ensuring successful ERP implementation the six major factors on clarity in goals and objectives behind the implementation, adequacy of user training, competency of the project implementation team, acceptance of changes brought about by the implementation and adequate vendor support and external consultant participation had a key role to play [1]. Similar studies conducted in china by creating interactive structural model have identified four critical factors on the funds support, department s participation; training and service of the supplier of ERP which influence s the system of ERP implementation most directly. The four factors above are critical factors which decide the ERP system is successful or not [2]. Based on the analysis of conditions of SMEs of China, the 6 CSFs of ERP implementation was suggested: top management support, great competence project team, right implementation scope, management program change, data accuracy, education and training in another study [3] In a study conducted in Malaysia by studying ERP Implementations in SME have identified 10 success factors on implementing team s teamwork and composition, Effective training of users, Open and honest communication,group structure, Other departments participation, Reasonable expectation with definite targets, Top management involvement, Cooperation between enterprise & Software Company, Project management, Effective decision - making [4] In Australia nine case studies were considered for investigation and four CSF s have identified and they are Success factors identified include: cross-functional team approaches, organizational experience of similar scale IT or organizational change projects, and deep understanding of all the key issues relating to an ERP implementation [5]. Similarly many studies have also focused on failure factors in ERP implementation. the failures are considered as stepping stones of success, the study of failed projects also has. These studies throw light on the failure factors in ERP projects. The of project management is studied using case study approach. The study has revealed that conflict among stake holders results in poor project management which intern results in ERP implementation failure [6]. Failure factors identified are identified in a study conducted on SME s in Australia and are listed as top-down or consultant driven implementations, IT department driven implementations, or implementations where the EFV is seen as a quick technological fix to problems within the operation of the firm, rather than as a strategic investment [6]. The best practices for ERP implementation in SMEs are investigated by using a problem driven approach by dividing implementation process into several components which will reflect the nature of ERP projects and makes them distinctive from other systems design. Methods critical issues relating to developing entrepreneurship for sustainable development of SMEs for its future extensions and added benefits are also addressed [7] The challenges in global ERP software implementation is studied.the challenges in implementation are identified as non uniform business practices in different countries, conflict of interests between various stake holders, Lack of experienced implementers in all countries, efficient uses of metanational advantages,. From this study it has been found that the benefits of ERP implementation are not same for all functions [8]. The major hurdle in ERP implementation is to find a match between the ERP system and an organization s business processes by appropriately customizing both the system and the organization. Framework for supporting management decision-making about Customization choices and the capabilities required to accomplish the match has been provided by various customization possibilities for ERP systems in a study [9]. The creation of value in organizations by knowledge management through implementation of ERP has been studied. In the study two stage approach is followed where Complete view of ERF' management Life cycle covering implementation framework is developed in first stage and practices of managing knowledge through the developed implementation framework is investigated in second stage. The findings of the study show that four knowledge domains reside in the firms that are human technology, management, and relational knowledge. Empirical evidence from these four cases showed that these knowledge domains were essential for business process improvement and ERP implementation. The specific knowledge management practices in each process of ERP implementation were also investigated and found that there are six phases of ERP adoption across its life cycle: i.e., plan, acquire, deploy, operate, optimize, and retire [10]. The implementation strategies for ERP to reduce the risk of failure have been studied. The simulation approach is used to understand the feasibility of a Big Bang strategy. Factors which can influence the choice of the more appropriate strategy are defined and specific ERP project planning in order to increase the chance of success is given. With various objectives, a number of methods, namely scoring, ranking, mathematical optimization, and multi-criteria decision analysis have been applied in the past to ERP projects (Wei, 2005) and a summary is given in Table1. We prefer the AHP as the methodology to evaluate the ERP implementation choices, as the major advantage of AHP over the other multi-criteria decision making methods is that AHP is designed to incorporate tangible as well as intangible factors, especially where the subjective judgments of different individuals constitute an important part of the decision process. Methods in practice Review of literature Scoring method Ranking approach Non-Linear programming model 0-1 Goal programming model AHP Method Table1: Review of Literature Methods in practice Review of literature This method according to Lucas & Moore (1976) is intuitive, but reflects opinions of decision makers[11]. Buss (19) proposed a ranking approach to compare computer projects and has found that this method has some limitations [12]. Santhanam and Kyparisis (1996) proposed a nonlinear programming model to optimize resource allocation allowing for the interaction of factors [13]. Badri and Davis (2001) presented the 0-1 Goal Programming Model. This method could not get the required attributes from an ERP system and hence got weakened [14]. Saaty (1980) discovered the AHP method[15] and was used by Schniederjans and Wilson (1991) in the ERP software selection process and was found useful[16]. 2

Nominal group technique (NGT) Teltumbde (2000) used NGT and AHP to select an ERP system and found AHP useful [17]. From the literature review it is evident that the success and failure factors for ERP implementation in SME s is studied extensively and other implementation issues like the quality issues are not explored much and there is a vast scope for study in this area. Specifically quality issues in ERP implementation are not explored in an Indian SME. Taking this into consideration the present study is aimed at analyzing quality issues in ERP implementation in Indian medium scale fertilizer industry and prioritizing the factors affecting ERP Implementation using AHP model. III. METHODOLOGY The quality issues that occurred during the implementation phase as well as the phase after which the system went live and began to be used by the trained personnel were recorded on site and maintained in a log book. Each of these issues was segregated on the basis of the module in which they originated and the frequency with which they occurred in these modules. The complete issue list was then considered for further analysis by means of a Pareto Chart using tools in Microsoft Excel.The major benefit of using this type of an analysis was to get the client to concentrate on just the vital few problems from the complete issue list that was recorded. This enabled a deep and thorough analysis of problems encountered in implementation so that a concentrated study might be carried out. The basic aim of this was to expose and analyze the key areas of the implementation process and procedure for various kinds of problems and other human factors which contributes to the end result of an unsuccessful implementation. A Cause and Effect analysis was carried out which highlighted the root causes which had a major contribution in the problems that both the client as well as the vendor faced during the course of the ERP implementation procedure. In this study, the analytical hierarchy process (AHP) is used as the methodology to Prioritizing the factors affecting ERP implementation on the basis of the priority value obtained by this method for each cell in the framework. The AHP provides a method to assign numerical values to subjective judgements on the relative of each element and then to synthesize the judgements to determine which elements have the highest priority. AHP is a method that advocates the comparison of two requirements at one moment. Please refer to Saaty (1980) for a detailed explanation on using the AHP method. It consists of the following steps: 4 5 Moderate + Strong 6 Strong + 7 8 9 Very strong, demonstrated Importance Very, very strong Extreme 2,4,6,8 Sublevels alternative is slightly preferred to the other. On the basis of experience and evaluation one alternative is favored strongly over the other. One alternative is favored strongly over the other; its dominance demonstrated in practice. The evidence on the basis of which one alternative is favored of the highest possible order of an affirmation. Source: Saaty, T. (1980): The Analytic Hierarchy Process, McGraw-Hill, New York, 54. In order to explain the mathematical model, we are supposed to start with certain assumptions. If n represents a number of criteria or alternatives, and C1, C2,, Cn a set of alternatives, the quantitative pairwise assessment of activities Ci, Cj is represented by n x n matrix A = ( aij ), (I, j = 1, 2,, n) where the elements aij are defined by the following rules (Saaty, 1980b): Rule1. If, then. This rule means that all the rows in the matrix are proportional to the first row and all of them are positive. Rule 2. If it is estimated that Ci is equally important as Cj, then it follows that aij = 1, aji = 1, and aii = 1 for each i. In this paper a model for ranking based on AHP method will be shown. Step 1: Choose the requirements to be prioritized Step 2: Set the requirements into the rows and columns of the n x n AHP matrix Step 3: Perform a pair-wise comparison of the requirements in the matrix according to a set of criteria Step 4: Sum the columns Step 5: Normalize the sum of rows Step 6: Calculate the row averages Intensity of 1 2 3 Table2: Saaty s scale Definition Equal Weak Moderate Importance Description Two alternatives contribute equally to the same goal On the basis of experience and evaluation one Figure 1. Structure of AHP model for ranking and comparison The reason for choosing this model is the fact it consists of a great number of criteria, not all of which of the same. In addition to this, a high quality computer system/software Super Decision has been developed. It is used in assisting the development of the model and enables a detailed sensitivity analysis of the final ranking list on the change of the values which are assessed subjectively. 3

In order to use this model for ranking, we need to determine the weights of the main criteria and sub-criteria, and then for each criteria at the bottom level of the hierarchy structure to define the intensities for the evaluation of the relevance. The weights of criteria and sub-criteria are calculated by the help of Super Decision software on the basis of the pairwise comparison of relative criteria and sub-criteria. For quantity criteria, the intensities are defined on the basis of the five-level scale of intensities (excellent, very good, good, satisfactory, weak), which have been derived on the basis of the range in which their values have fluctuated. IV. FINDINGS The major root causes were classified into four subheadings namely, the CRP, the top management, training and finally the data collection. The main effect was the challenge faced in the implementation of the ERP system. From earlier analysis by means of the Pareto Chart, it was found that the functional modules of Sales &Distribution and Financial-Costing posed majority of the problems. Hence, these problems needed to be addressed through a more precise tool which will bring out the exact areas of shortcomings in the whole system The 8 factors affecting the ERP implementation are determine. Choosing the right option is a difficult task as it requires consensus among the ERP team and top management. This indicates the need for application of the AHP to the framework. The results obtained using AHP are tabulated in Table and Figure. Sl. Factors No 1 Adequate and Correct Data 2 Training and Testing 3 Never run Parallel sysytem 4 Conference Room Pilot 5 Employee Retention 6 Customosa tion 7 Clearity In mngt objective 8 stakeholders Identificatio n Table 4: Null Alternative 95.00% CI μ 0=04 μ a 04 95 to 14 μ 0=37 μ a 37 33 to 42 μ 0=55 μ 0 55 53 to 58 μ 0= μ 0 82 to μ 0= μ 0 to μ 0= μ 0 82 to μ 0=09 μ 0 09 00 to 19 μ 0=46 μ 0 46 32 to 50 Table 5: t Prob Result 37 0.894 Null 41 0.891 Null 0.509 0.623 Null -.011 0.991 Null -5 0.958 Null 38 0.817 Null Hypothesi s Accepted 18 0.908 Null Hypothesi s Accepted. 19 0.9 Null Hypothesi s Accepted. Factors Priority value Percentage % Priority /Rank Adequate and Correct Data 04 30.4 1 Training and Testing 37 13.7 3 Never run Parallel sysytem 55 15.5 2 Conference Room Pilot 8.4 6 Employee Retention 8.5 5 Customosation 8.3 7 Clearity In mngt objective 09 10.9 4 stakeholders Identification 46 4.6 8 Table: Priority table6 V. RESULTS AND ANALYSIS Factors Adequate and Correct Data Training and Testing Never run Parallel system Conference Room Pilot Employee Retention Customosati on Clearity In managemen t objective Stakeholder s Identificatio n Figure 2. Table 3: Response 1 2 3 4 5 6 7 8 9 10 92 36 60 87 89 07 01 46 50 80 12 90 39 60 82 15 45 02 40 58 81 06 94 40 56 82 12 11 43 51 81 81 03 47 96 34 57 14 45 13 39 53 87 95 47 26 30 54 87 91 23 27 57 82 38 05 After the study the following suggestion were given to the company to assist them in re implementation. 1. Adequate and correct data should be provided- the For Multiplex to perform a make-over to the new ERP system, data had to be collected from the distributed Tally 7.0 Servers, had to be reconciled, mapped into the SAP R/3 System in its standard format and finally the data had to be uploaded into the SAP system. The problem on-site was about 40% inappropriate data was provided by each of the different branches hence making reconciliation a major issue. Hence, strong management direction is needed for the managers at each of the branches so that adequate and appropriate data is duly provided. 2. Never run parallel systems- Multiplex had been using Tally 7.0 prior to the installation of SAP R/3. The employees had been using the same for years. Therefore when issues began to crop up after implementation of SAP R/3 in Finance module, sales and distribution module was completely ignored, they shifted work with these modules back to the old system. This hampered the proper integration of organization data and led to data mismatch in other modules as well. As a result, support system provided by the vendor became obsolete and difficult to implement. Hence, use of parallel systems should be avoided outright. 3. Training and testing of the system should be done properly- Training by the SAP Consultants, that is, the vendor is provided as part of the implementation procedure to only a 30% group of people from the client s side known as the Core Team. This core team in turn trains a rest of people who are actually responsible for day-to-day transactions called the End Users. It was observed that the 50% second leg of training which is provided to the end users was not carried out mainly due to lack of computer literacy, not will to accept the 4

responsibility This triggered a strong resistance to change for the new system being installed and caused reduction in employee motivation. 4. Clarity in management objectives and expectations from the ERP System- The top management at Multiplex had not defined its objectives of implementing the new SAP R/3 package clearly to the vendor and had expectations way beyond its initial scope. This led to a disbelief of the system s power to integrate the company s functions. According to the vendor, Multiplex management expected a quick return on investment (ROI) which was not practical since it takes around three to four months to notice any significant returns. Hence, top management should be patient with the new system and any fear of failure should be done with for a successful running system. 5. Employee retention programs should be practiced- It was observed that after the completion of SAP R/3 training provided to the Multiplex staff and within some days of the system going live, six of the thirty SAP R/3 trainees from the organization quit the company causing great losses to Multiplex in the form of shortage of key resources i.e. trained staff. This was a big percentage of employee attrition rate and it is not possible for a company to hold back any of its employees even with the most stringent contract. To counter this situation, Multiplex should ask the employee to inform well in advance before leaving and train an alternative person who can take over his job. As a counter measure, Multiplex should offer its employees Attractive salaries, Stock options, Challenging and a comfortable work environment. 6. CRP (Conference Room Pilot) is a very important part of software testing and should not be neglected- We recommend a conference room pilot approach where computer work stations are set up in a room to represent each of the major tasks of customer service /order entry, planning, goods-in, stores and finance. A simplified data set is loaded and the company operations run through. The data is gradually increased as first the project team, then managers and finally users get more familiar with the software. This is conducted just before the ERP becomes fully functional in the organization.crp is a major part in software testing and is a big step towards a successful Go-live period for the organization since the end users not only strive to perform as per training but also mistakes in operating the system can be pointed out by the experts in a live environment. 7. Customization should be less than 30%- Customization Services involves any modifications or extensions that change how the out-of-the-box ERP system works. Customizing an ERP package can be very expensive and complicated. Some ERP packages have very generic features, such that customization occurs in most implementations. Customization work is usually undertaken as "changes requested beforehand" software development on a time and materials basis. But ideally, experts in the ERP implementation field have suggested that customization should be less than 30%. The level of customization in the case of Multiplex exceeded beyond this and posed a great deal of problems when key applications were run and found to be not working as they were intended to. 8. Stakeholders shall be identified in the initial phase including customers and vendors- Stakeholders are all those who are directly or indirectly affected by a company implementing any new ERP system be it organizations like those of the supplier as well as the vendors. A failure to identify the stakeholders gives the implementing company a major setback when the concerned people or organizations work against the new system. Hence proper identification of all stakeholders to be done. REFERENCES [1.] Parijat Upadhyay et al, An explorative study to identify the Critical Success Factors for ERP implementation in Indian small and medium scale enterprises, 978-0-7695-3513-5/08 $25.00 2008 IEEE, DOI 1109/ICIT.2008.66 pp 295-299 [2.] Ranzhe Jing et al, A Study on Critical Success Factors in ERP Systems implementation, 1-42-08-7/07/$20 2007 IEEE. 2007 [3.] Yuanqiang Xia et al, The ERP Implementation of 8ME in China, 978-1-42-3662-0/09/$25.00 2009 IEEE pp 135-140 [4.] Ali Noudoostbeni, To investigate the Success and Failure Factors of ERP implementation within Malaysian Small and Medium Enterprises, 978-0-7695-3595-1/09 $25.00 2008 IEEE, 2009, pp 157-160 [5.] Anthony Marsh, The implementation of enterprise resource Planning systems in small-medium Manufacturing enterprises in south-east Queensland: a case study approach, ICMIT, 2000,0-7803-6652-2/2000/$10 2000IEEE pp 592-597 [6.] Charlie C. Chen, Chuck C. H. Law, and Samuel C. Yang, Managing ERP Implementation Failure: A Project Management Perspective, IEEE, 2009,Vol 56, pp 157-170 [7.] Rashmi Jha, M. N. Hoda, A. K. Saini, Implementing Best Practices in ERP for Small & Medium Enterprises, 978-1-42-2972-1/08/$25.00 2008 IEEEpp 1-5 [8.] Saumyendu Ghosh, Challenges on global implementation of ERP software, IEE, 20029-7803-73-5/02/$17.00 2002 IEEE pp 101-106 [9.] Wenhong Luo and Diane M. Strong, A Framework for Evaluating ERP Implementation Choices, IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT, VOL. 51, NO. 3, AUGUST 2004 Vol 51, pp 322-333 [10.] Phasit Kanjanasanpetch, Managing knowledge in Enterprise Resource Planning (ERP) Implementation, 0-7803-8150-5/ 03/$17.00 02003 IEEE, pp 30-35. [11.] Lucas, H.C., & Moore Jr., J.R. (1976). A Multiplecriterion scoring approach to information system project selection. Infor., 14(1), 1-12. [12.] Buss, M.D.J. (19). How to rank computer projects. Harvard Business Review, 61(1), 118-125. [13.] Santhanam, R., & Kyparisis, G.J. (1996). A decision model for interdependent information system project selection. European Journal of Operational Research, 89, 380-399. [14.] Badri, M.A., & Davis, D. (2001). A comprehensive 0-1 goal programming model for project selection. International Journal of Project Management, 19, 243-252. [15.] Saaty, T. L. (1980). The analytic hierarchy process (AHP). New York: McGraw-Hill. [16.] Schniederjans, M. J., & Wilson, R. L. (1991). Using the analytic hierarchy process and goal programming for information system project selection.information & Management, 20, 333-342. [17.] Teltumbde, A. (2000). A framework for evaluating ERP projects. International Journal of Production Research, 38(17), 4507-4520. [18.] Parijat Upadhyay, Pranab K Dan, An explorative study to identify the Critical Success Factors for ERP implementation in Indian small and medium scale enterprises 978-0-7695-3513-5/08 $25.00 2008 IEEE [19.] Ali Noudoostbeni1, Norizan Mohd Yasin To investigate the Success and Failure Factors of ERP implementation within Malaysian Small and Medium Enterprises 978-0-7695-3595-1/09 $25.00 2009 IEEE [20.] Antony Marsh The Implementation of Enterprise Resource Planning Systems in Small-Medium Manufacturing Enterprises in South-East Queensland: A Case Study 5

Approach ICMIT 2000 0-7803-6652-2/2000/ $1002I0E0E0E [21.] Al-Mashari, M. and Zairi, M. (2000), ``Information and business process equality: the case of SAPR/3 implementation, Electronic Journal on Information Systems in Developing Countries,Vol. 2 [22.] Gable, G. and Stewart, G. (1999), ``SAP R/3 implementation issues for small to medium enterprises, 30th DSI Proceedings, 20-23 November, pp. 779-81. [23.] Zairi, M. and Sinclair, D. (1995) ``Business process re-engineering and process management: a survey of current practice and future trends in integrated management, Work Study, Vol., pp. 6-13. [24.] Markus, M. (19), ``Power, politics, and MIS implementation, Communications of the ACM, Vol.26 No. 6, pp. 430-. [25.] Aaker, D. (1992), Developing Business Strategies, John Wiley, New York, NY. [26.] Amoako-Gyampah, K. (1999), ``User involvement,ease of use, perceived usefulness, and behavioral intention: a test of the enhanced TAM in ERP implementation environment, 30th DSI Proceedings, 20-23 November, pp. 805-7. 6