FINAL PROJECT ADMS 4300 FALL 2011 Decision Support System and Managerial Decision Making Prof. Hassan Qudrat-Ullah Due Date: December 8, 2011 By: Farazeh Khalid Mian Abhishek Sahi Table of Contents SECTION NUMBER| SECTION NAME| PAGE NUMBER| | Abstract| 3| 11. 11. 2| IntroductionWhat is decision making? What is a Decision Support System? | 3, 44, 5, 6| 22. 12. 22. 3| Literature Review DSS in the business environmentImportant attributes of the Decision Support SystemCapabilities offered by DSS| 77, 8, 910| 33. 13. 23. 33. 43. 4. 13. 4. | Discussion and AnalysisHow are Decision Support Systems and managerial decision-making linked? Managers and Decision Support SystemsDSS implementationA real-life case on how a company’s problem was transformed into a DSSTransforming the GSK problem into a Decision Support System — a conceptual viewSimulation Model| 11, 1213, 1414, 15, 1617, 1818, 1920, 21, 22| 4| Conclusion| 22, 23| | References| 24, 25| Abstract: In today’s world, business is rapidly evolving and there’s a greater need to make better-informed decisions using the fast flowing and wide variety of information available.

This paper focuses on how Decision Support Systems (DSS) contribute positively to organizations and how managers across several industries use a DSS to assist in tactical decision-making. We discuss what a DSS is, and it plays a significant role in managerial decision-making. Computer-based information systems play a critical role in providing significant support for unstructured and disoriented problems. We discuss how DSS allows organizations to gain advantage in their daily processes and allows them to tackle their problems in an organized and informed manner.

We discuss and analyze an example on Glaxo Smith Kline (GSK), an organization that transformed a GSK problem into a decision support system framework by creating a model outline for the proposed DSS, and then implementing the DSS. We conclude that Decision Support System is a powerful support tool that significantly contributes to boosting individuals and an organization’s productivity, given it is planned and integrated effectively. Keywords: DSS, Decision-making, Managers, Glaxo Smith Kline 1- Introduction: 1. 1 - What is decision making?

A decision is a mental process that can make or break an organization’s success. Decision-making involves analyzing and calculating the mathematical probabilities of success and failure of the different alternatives in a given situation, and then taking action based on one of the alternatives. A decision is the process of taking a stand for something, or a position or belief or conclusion reached after deliberation of alternatives and options. The importance of a right or a wrong decision in an organization cannot be stressed on enough.

A decision could lead to outright success, or failure. The consequences are far-reaching for the rest of the organization’s life. Figure 1 illustrates the decision making process. Fig 1: The decision-making process 1. 2 - What is a Decision Support System? Decision Support Systems (DSS) are a branch of computer-based information systems that’s main purpose is to support decision-making activities in an organization. Computer- based information systems include management information systems, transacting processing systems and office automation systems.

Support systems that are used to help the management level in an organization include DSS, executive information systems, dashboards and expert systems. Model and data management in DSS are entangled fields: data is a fact that is produced as output from observing occurrences such as daily production quantity, daily sales quantity and inventory levels of products. The database is the component that integrates interrelated files; and the database management system uses the database to produce results the user wants to generate using queries and filters.

Although data management in DSS is required for a functioning support system, it’s not sufficient to handle design and choice stages of decision-making processes. In order for a DSS to be successful, it must be able to handle activities such as projection, deduction, analysis, creation of alternatives, comparison of alternatives, optimization and simulation (Sprague and Carlson 1982). To perform these crucial tasks, DSS employs several kinds of management science/operations research (MS/OR) models, which include linear programming, integer programming, network models, goal programming, simulation, statistical models and spreadsheet modelling.

These models are stored in the model base and the model-based management systems are computer programs that are used as a component of a DSS generator to build, restructure and update models. A DSS consists of three fundamental components: 1. The model 2. Database 3. User interface The interface system allows users access to i. The data-subsystem which includes the a) database, b) database management software ii. The model sub-system which includes the a) Model base, b) Model base management software Illustrated in the diagram below) (Decision Support Systems, Sean B. EOM, 2001) 2- Literature Review: 2. 1 - DSS in the business environment: In order for DSS to be useful and successful implementations, they need to be maintained and structured well. It is also important that the data processing system should be organized as that provides data that is necessary for DSS. In order to set up and to maintain a DSS system, an organization must invest and regulate its financial and human resources into it.

The system is structured in a way that there is a central computer, which connects to computers in all other departments. The departments must have sufficient understanding of the system as well, which is why organizations must also invest in education and training programs to explain the features of the system. 2. 2 - Important attributes of the Decision Support System: 1) The most imperative deliberation is the Decision Support System’s advantage of simple usage – its aptitude to help people with no technical know-how to handle it directly.

The sole biggest and most permanent flaw with computers has been their rigidity, their failure to let the user who is in dire need of the data to handle the computer on his own. 2) The capability to access information must not only be limited to a fraction of a firm or to merely definite administrative or qualified groups. In its place, the resource should be supplied to all of the members and fraction of a firm wanting it devoid of extensive access. The authority of sophisticated Distributed Processing System will go unexploited as they usually have historically. ) The ultimate Decision Support System in prominent disparity to the preceding technique of devising software must not be a ‘system’ at all in the stern logic of the term. In fact, it must be an extremely adaptive decision support generator that can effortlessly be used by experts to hastily devise data support examplar matched to every particular decision-making assignment. This adaptive instrument should permit rapid design modifications if the first-hand design does not narrowly suit a user’s information seeking manner or requirements. ) To sufficiently support the human constituent, this extremely adaptive support ability should be able to give access to functioning information and also to summary information that previously has been completed by application programs devised for other definite functional errands. At the same level significance, this instrument should give the expert the access to a firm’s raw data and it must allow the access to be completed in a single step using a solitary simple process or instruction and without having to re-key non summary data. ) The firms require to access first-hand information occasionally because efficiency is connected to how fit the first-hand information is structured in the system; the Decision Support Generator must be able to line with the right DBMS. It must also be able to access benchmark ‘flat’ files circuitously by means of the authority of the host computer to assist both the user interface and information access without amending current files. 6) The Decision Support Generator must let the person choose whether data must be screened on the CRT screen for instant exercise or whether it must be printed for later implementation.

The finest method to achieve such supple information portrayal is via a work place. The administration or expert data workstation would integrate a keyboard, display screen and an interface to a printer, which can print all from straight text to graphics like pie charts, bar charts and line charts. 7) The support instrument should line with numerous dissimilar systems and abilities, it should be well-suited to all of them, the instrument should give persons a solitary easy to use language to access, manoeuvre and portray information in a manner that will support the end-user at its finest. ) To assist editing and manoeuvring presented information, the decision support generator must preferably be able to line with word processing software. With this ability, the DSS develops into the crucial link amid information dispensation and office mechanization, incorporating both uses in an easily used, clear-cut, tremendously influential structure. Decision making characteristics in the Global Business Environment: Business Strategy| Decision Making Characteristics| Multinational (decentralized federation)| Decision making decentralized to subsidiaries, Informal relationships between head quarters and ubsidiaries| International (coordinated federation)| More vital decision and knowledge in general developed at head quarters and transferred to subsidiaries| Global (centralized federation)| Decisions made at the center knowledge developed and retained at the center| Transnational (integrated network)| Decision making and knowledge generation distributed among units| “Decision Support System and Managerial Decision Making”, International Journal of Knowledge and Research in Management & E-Commerce Vol. 1, Issue 1, Dr. VVR Raman, Dr.

Veena Tewari, January 2011 2. 3 - Capabilities offered by DSS: A DSS offers features that fill an organization’s several needs. 1. It has the capability of assisting decision-making in situations and problems, which are ambiguous and ill structured. The amount of data that needs to be accessed by such a situation is handled by the DSS, providing the decision maker with a clearer picture, to a previously chaotic situation. 2. They allow for clearer and well informed decisions as the decision is based on information, as well as human judgment 3.

Decision support systems are able to provide data as immediately as required in a time sensitive situation 4. DSS can function in the ad hoc mode which suits itself according to the needs of the user as opposed to operating in a broader scheduled manner a management reporting system provides 5. They do not provide generic results; they are able to adapt to an individual’s particular decision making style, making it flexible and more comfortable to use 6. They have a particular feature that allows group decision-making, referred to as GDSS. 7.

They smooth the way for implementation of decisions that are made across department limits 8. DSS helps managers to develop a more comprehensive understand of their business activities through the development and integration of models 9. Supports several stages of the decision-making process 3- Discussion and Analysis: 3. 1 - How are Decision Support Systems and managerial decision-making linked? In this paper, we discuss how DSS contributes positively to support managerial decision- making. In today’s world, to be an effective manager, automated support has almost become a requirement.

Decisions must be made more quickly than before and the process of identifying problems and solutions has become an increasingly difficult task; it seems impossible to rely on human judgement alone. A DSS is designed to specifically integrate the brain of a human with that of a computer’s and with this abundance of data and information; it allows managers to look at all aspects of the problem, and to respond in accordance. DSS allows managers to react and to readjust to the continually transforming business environment; and it allows an organization to have a competitive advantage against its competitors.

A DSS’s strength lies in combining the human skills of experts with the power and capabilities of a computer-based information system to give managers the accurate and efficient reporting, statistical and modeling capabilities, in-depth analytical reports, high volumes of data and a variety of Graphic User Interface (GUI) to represent this data in different views in accordance to the preference of the user. Decision support systems are useful to managers across many industries and occupations; it is not restrained to a business organization. Figure 2 shows the many applications of DSS:

Geodata Analysis Display System(GADS)| Geographical resource application and analysis; applications include sales force territories, police beat redesign, designing school boundaries| Portfolio Management System(PMS)| Portfolio investment management| Industrial Relations Information System(IRIS)| Ad hoc access to employee data for analysis of productivity and resource allocation| Interactive Financial Planning System(IFPS)| Financial modeling, including mergers and acquisitions, new product analysis, facilities planning and pricing analysis| Interactive Support System for Policy Analysts(ISSPA)| Policy analysis in state government; simulations, reporting and ad hoc modeling| Interactive Marketing System(IMS)| Media analysis of large consumer database; plan strategies for advertising| Computer-Assisted Underwriting System at Equitable(CAUSE)| Calculate and track group insurance policy and renewals| An Analytical Information Management System(AAIMS)| Analysis of time series data on airline industry operations (database contains airlines’ reports to Civil Aeronautical Board)| Figure 2: Applications of the DSS Source: “Decision Support Systems and Managerial Productivity Analysis”, Peter G. W. Keen, October 1980, CISR No 60 3. 2 - Managers and Decision Support Systems:

A manager is constantly surrounded by a plethora of problems: enormous number of matters that require immediate attention, some possibly at the same time, along with new problems constantly materializing. A lucrative manager must also need to focus on networking and therefore maintain an exceptionally large number of contacts, both inside and outside of the organization. The types of information that is provided by decision support systems has always been a requirement and due to a this, demand for such systems has increased. In the modern day and age, there are easily available methods for storing and maintaining a tremendous amount of data. The ease in using software packages that possess the tasks of decision support systems allows many organizations to implement these systems effectively.

Data processing software allowed managers to stay on top of things and gave them the ability to influence others outside the organization as well. Since managers are in need of maintaining a constant database of their data, the ability to come to a platform where all this data would be ‘combined’, was a greatly beneficial step for organizations; helping the managers make decisions better and quicker. A manager’s job poses great challenges and with those challenges come great threats, as a ‘wrong’ decision could lead the organization into danger. It is believed that a manger’s job while making decisions is well defined: analyzing relevant information, studying the alternatives and logically picking the alternative that maximizes benefit and minimizes uncertainty.

However, this job is never that simple or easy. A manager plays three types of roles: 1. Interpersonal roles: Face-to-face communication, assisted by electronic means. Information systems assist in serving the information and providing a platform for communication. 2. Decision roles: This is the critical aspect, that requires a manager to be able to gather and use all resources effectively, as well as being able to handle disturbances 3. Problem solver: Being able to identify problems is a crucial step to decision making; a manager should be able to analyze what kind of attention the problem demands and in turn, study the alternatives effectively 3. 3 - DSS implementation:

In order for a DSS to be effectively implemented in an organization, there must be clear-cut goals that are worked towards during the process of creating a system. In attempt to enhance efficiency, a foundation must be established and it is important to determine 1. How the job is currently done 2. How and where can performance be improved It is from the analysis of these factors that a DSS can be built according to the needs of the organization: 3. What must happen for the manager to use a DSS that allows room for improvement? 4. What is the cost of implementation? One of the difficult tasks in improving productivity in intricate tasks is simply that they are quite complex. Thus the manager must determine the order in which activities are supposed to be collaborated.

Due to the nature of the situation, it is almost unattainable for the analyst to fully understand and then to be able to set out the specifications for a DSS, and for the manager to explain it in detail enough for the analyst to understand. David Ness (a professor at MIT) uses the phrase “middle-out” design to capture the way in which a prototype can find a solution to this predicament. Middle-out is where the entire structure is set out in advance and where separate components of the system are developed and collaborated at a later stage. This point is argued and is support by several DSS case studies that the DSS builder must put together a definitive and operational system to which the manager can respond.

From there on, evolving the DSS is done through exploring the system, and thus modifying and extending it as required. Meador, Guyote and Rosenfeld (1986) suggested a four-stage process for developing a DSS: decision support analysis, DSS software evaluation and selection, prototype development and operational deployment and support (M. C. ER, 1988). It is illustrated in the diagram below: Source: “Decision Support Systems: A summary, problems, and future trends”, M. C. ER, 1988, Elsevier Science Publishers B. V. (North-Holland) 3. 4 - A real-life case on how a company’s problem was transformed into a DSS: GlaxoSmithKline (GSK) is a pharmaceutical company headquartered in the UK with operations based in the U. S.

It is the second largest pharmaceutical company in the world with $35 billion in sales, which accounts for 7% of global pharmaceutical sales. The North American headquarters are in Pittsburgh, PA, and Parsippany, NJ. About $7 billion of GSK's revenues are derived from Consumer healthcare products, which include over-the-counter (OTC) medicines, oral care products and nutritional healthcare drinks, all of which are among the market leaders. In the OTC segment, popular products include Nicorette, Tums, and Contac. The oral care products consist of Aquafresh, Sensodyne, and Polident brand names. GSK also provides nutritional drinks such as Horlicks.

Its major competitors in the consumer healthcare markets include Colgate–Palmolive, Johnson ; Johnson, Pfizer, Procter ; Gamble, Unilever, and Wyeth. Thirty-six percent of OTC sales are in North America. They focus on 51 brand groups, ten of which have annual sales greater than $100 million. There are altogether 1200 product types. Annually many new and improved products are introduced, and they replace 20–30% of the existing ones. The GSK supply chain includes four plants, thirty contractors, two co-packing facilities, and four regional distribution centers (RDCs). There are 400 customer warehouses and 25,000 retailer ‘ship-to’ locations. Annually 80,000 customer orders are received, and 20 million cases of products are shipped. Average inventory valuation is $115 million.

In GSK, inventory level is measured in Weeks Forward Coverage (WFC), which is the number of weeks the inventory can cover and is based on the 6- month forecasted demand. Current WFC is 17 weeks. GSK's dollar fill target is set at 96. 6%, whereas that of the new item launches is aimed at 100%. Although fairly profitable, compared with competitors, GSK consistently maintains a higher inventory level. Our goal in this project was to develop a model that would determine the best level of safety stock and the accompanying WFC at the SKU level for GSK's products, such as Nicorette and Aquafresh. Determining optimal inventory levels in the multiechelon supply chain setting is a complex problem that requires trade-off of various stochastic variables.

While it is desirable to provide a high service level (SL) to customers, it is also important to keep inventory levels low to save on cost. A high inventory level is unhealthy, because it represents an investment with very low return. Since inventory turnover is the cost of goods sold divided by the average inventory level, a low turnover implies excess stock, cash tie-up, sluggish sales, ineffective buying, or vulnerability to falling prices. 3. 4. 1 - Transforming the GSK problem into a Decision Support System — a conceptual view: They started with a big picture of “converting” GSK's problem into a DSS: the inputs, the model, and the outputs. Fig. 3 gives an overview of the planned DSS. At this stage of their project, the main concerns were the input and the output.

They needed to determine the system inputs so that they could request the data from GSK's management. They also needed to articulate the outputs in terms of GSK's needs (“answers” management can use and rely upon). A reasonable set of input variables seemed to be: historical forecast and demand, month-end inventory, the production batch size, and the lead time. They decided that the outputs from the DSS would be a point estimate of the WFC (which is equivalent to safety stock in traditional inventory management terminology) at the SKU level. In addition to providing this WFC, they wanted to give management the flexibility to study through a simulation model the effects of WFC on the service level. Fig. 3. Model outline for proposed DSS.

As they designed the DSS, they worked closely with several managers from GSK, including the Vice-President of Customer Supply, the Business Unit Planning Manager and Site Coordinator, the Director of Customer Service, the Manager of Planning Systems & Forecasting Relationships, and the Director of Distribution. They all provided valuable and timely feedback; the Business Unit Planning Manager and Site Coordinator was especially active and responsive to their needs. Top management were very interested and involved in the project and closely monitored the steps taken to design and build the DSS. Every key procedure and decision were explained to the managers, and the design process was transparent to them. Because of the commitment and involvement of these managers from GSK, there was a high degree of trust in the development team. 3. 4. 2 - Simulation Model:

The inventory model contains heuristics, and the heuristics are imbedded in the simulation. There are several reasons why a simulation is needed: (1) Many of the product demands in GSK are seasonal. Average demand and standard deviation of demand vary greatly depending on the planning horizon. Due to the stochastic nature, dynamic data has to be replicated through simulation. Thus simulation is chosen for its suitability for generating weekly demand. (2) The target inventory level, needed by the spreadsheet modeling, is not constant. It changes with the forecast over time. Such information needed by the heuristic model is best supplied by simulation.

Note that the simulation is not to replace the heuristic inventory model. Instead the inventory model uses simulation to generate valuable information that helps decision making. (3) To answer what-if questions under different safety stock levels, a simulation model is necessary. By experimenting with various levels of SS, one can empirically learn if the recommended SS performs better than the other alternatives, both in terms of inventory and service levels. Since GSK's demand and replenishment characteristics are dynamic, and most input data vary with time and require continuous monitoring, it made the most sense to build an add-in or a stand-alone program to address this situation.

The standalone spreadsheet simulation module is described below. Fig. 3. Outline of the simulation logic. Table 1 Nomenclature for the simulation logic of Fig. 3 4- Conclusion: Within the precedent a number of years, computers have been used progressively more in fields of financial management, manufacturing study, temporary planning and geographical study. Currently in commerce world, computers are taken advantage for decision-making process as a Decision Support System and management tool. Decision Support System is a kind of administrative data system whose primary intention is to assist a human decision maker during the course of his/her taking a stand to make a decision.

The power of Decision Support System lies in assisting decision taking in scenarios where both human analysis and decision and the authority of the computer are essential. Decision Support Systems principally shore up for tactical, intellectual and functional planning. Accurately developed and incorporated, the Decision Support System becomes an extremely strong assistance apparatus that boosts the efficiency of proficient people at all business ranks in every division. It can efficiently expand the business’ present work force by plummeting its burden thereby, escalating efficiency. And with the current and latest technological know-how and status of the art software equipment, it can bring humans even closer to fill the gap and functional accomplishment worlds.

These attributes can recommend current businesses suppressed more than ever in the past, to exploit effectiveness while plummeting costs, extraordinary advantages in the consumption and administration of both their human and computer capital. Therefore, we conclude along with supportive evidence that DSS’s are highly beneficial to organizations, providing them with competitive edge and a wider frame of approach when it comes to decision-making. A decision holds a lot at stake for an organization, potentially its success or failure, and a DSS investment has proven to go a long way. References: 1. Jennifer Shang, Pandu R. Tadikamalla, Laurie J. Kirsch, Lawrence Brown, A decision support system for managing inventory at GlaxoSmithKline, Decision Support Systems, Volume 46, Issue 1, December 2008, Pages 1-13, ISSN 0167-9236, 10. 1016/j. dss. 2008. 04. 004. (http://www. sciencedirect. om/science/article/pii/S016792360800078X) 2. M. C. Er. 1988. Decision support systems: a summary, problems, and future trends. Decis. Support Syst. 4, 3 (September 1988), 355-363. DOI=10. 1016/0167-9236 (88) 90022-X http://dx. doi. org/10. 1016/0167-9236(88)90022-X 3. Information Systems in Management Science: Decision Support Systems: The New Technology of Decision Making? Andrew Vazsonyi, Interfaces, Vol. 9, No. 1 (Nov. , 1978), pp. 72-77 Published by: INFORMS Article Stable URL: http://www. jstor. org/stable/25059694 4. Nolan, Dr. Richard L. , Decision support systems: Managerial tools enhance decision-making, Tymshare, Inc. , 1977 5. Veena, Tewari, and Raman Dr. VVR. Decision Support System and Managerial Decision Making. " . International Journal of Knowledge and Research in Management & E-Commerce Vol. 1, Issue 1, January 2011. Web. 7 Dec 2011. 6. Sean B. Eom, "Decision Support Systems," International Encyclopedia of Business and Management, 2nd Edition, Edited by Malcolm Warner, International Thomson Business Publishing Co. , London, London, England, 2001, forthcoming. 7. F. J. Radermacher, Decision support systems: Scope and potential, Decision Support Systems, Volume 12, Issues 4-5, November 1994, Pages 257-265, ISSN 0167-9236, 10. 1016/0167-9236 (94) 90044-2. (http://www. sciencedirect. com/science/article/pii/0167923694900442)

Keywords: Artificial intelligence; Biases; Decision support; Decision theory; Modelling; Optimization 8. Salvatore T. March, Alan R. Hevner. Integrated decision support systems: A data warehousing perspective. Decision Support Systems, 2007: 1031~1043 9. Power, D, and R Sharda. “Model-driven decision support systems: Concepts and research directions. ” Decision Support Systems 43. 3 (2007): 1044-1061. 10. J. P. Shim, Merrill Warkentin, James F. Courtney, Daniel J. Power, Ramesh Sharda, and Christer Carlsson. 2002. Past, present, and future of decision support technology. Decis. Support Syst. 33, 2 (June 2002), 111-126. DOI=10. 1016/S0167- 9236(01)00139-7 http://dx. doi. org/10. 1016/S0167-9236(01)00139-7