Decision Support System in Management Decision-Making - Literature review Example

Decision Support System in Management decision making By Name: Module: Instructor: Date: Introduction Explosion in technology since early 1990s has resulted to managerial shift in decision making processes. Advertising industry is reaping from change in customer behaviour trends as most of the world population adopts the emerging technologies. However, advertising firms face enormous task in managerial decision making strategies to cope with scenario problems and intense competition in the industry as well as achieving targets. For a number of years research has been going on to develop re-usable and shareable scenario problem-solving components to support in decision making (Knoo & Forgionne .n.d.). Management forms and integral facet in an organization, it is actually the principal management component where complex decision making processes are developed. Decisions made in a business set up are categorized as structured, unstructured or semi-structured. Structured decisions, often referred to as ‘programmed’ are mostly routine, procedural and within specific policy outlines. While, unstructured decisions involve unique or new problems and simulations scenarios which do not follow routine or programmed decisions. Likewise semi-structured decisions involve scenario problems but encompass both unstructured and structure components. Depending on the size of an organization, decisions are made by teams, groups, or individuals (see Figure 1.0) (Knoo & Forgionne .n.d.). Figure 1.0 shows organizational structure and decision making process. Where: A = Information B = Decision However, organizational management diversity in terms of experience and knowledge often leads to differences in decision making process depending on individual reaction to problem scenarios. To harmonize the process of decision making, the following steps are essential; Intelligence, Design, Choice, Decision, and Implementation. These steps are complimented by strategies such as, Optimization, Satisfying, Elimination based on facts, Incrementalism, mixed scanning, and Analytical Hierarchy process. Similarly, the strategies must be grafted within a specific model. Some of the common model types applied in decision making includes Deterministic, Stochastic, Simulation, and Domain-specific. In this paper stochastic model has been applied through regression and queuing theory (Knoo & Forgionne .n.d.). In Stochastic information Technology forms the cornerstone for formulation of decision making process, this is attributed to IT infrastructures which facilitate communication and team interaction, data assimilation and filtering, offering support in problem recognition, problem solving and cohesively putting together the results in a compact package. Among the most common applications of Stochastic models in managerial decision support are Geographic Information System (GIS), Data Mining and Group Support system. However, there are other systems which are based on same framework but having a more specific target. This includes, Business Intelligence Systems, Knowledge Management Systems, Artificial intelligence, Virtual Reality, Neural Networks and Expert Systems (Wijnhoven .n.d). Decision Support System (DSS) This is a computer based system, well designed to assist decision makers to find solutions to arising unstructured problems (Knoo & Forgionne .n.d.). The system is usually interactive and incorporates data and models. The data is gathered from data warehouse or transaction processing systems (See Figure 2.0) Figure 2.0 is a schematic representation of a Decision Support System A system shown in Figure 2.0 above is further classified as 1.data-oriented, where it provides tools for analysis and manipulation of data. 2. Model-based, this supports certain mathematical model for supporting the process of decision making (Knoo & Forgionne .n.d.). Discussion and Results This paper considered the later model in analysis of market factors encountered in advertising firms. Regression analysis was used to identify target segments among the population in order to assist in launching advertising campaigns. Within regression, Analysis of Variance (ANOVA) results are used to compare the two modes of advertising strategies, this are 1. Glazing Advertising and, 2. Conservatory Advertising. Regression out come is defined by Variable Y while variable X is the predictor (forecast value), and Standard error as the controls for validity and variability. This is the p-value of the model. It tests whether R2is different from 0. Usually we need a p-value lower than 0.05 to show a statistically significant relationship between X and Y. R-square shows the amount of variance of Y explained by X. In this case income stream explains 0.5% of the variance in Ads. The forecasts were determined from sales revenue data and dollars spent on running Ads. Two-tail p-values test the hypothesis that each coefficient is different from 0. To reject this, the p-value has to be lower than 0.05. In this case, Population is statistically significant in explaining revenue returns (Wijnhoven .n.d). The t-values test the hypothesis that the coefficient is different from 0. To reject this, you need a t-value greater than 1.96 (for 95% confidence). You can get the t-values by dividing the coefficient by its standard error. The t-values also show the importance of a variable in the model. Followed by linear regression analysis, from the out put results (see Appendices), it is reasonable to infer that advertising played part in driving sales revenue, however, the relationship strength measure between the variables were needed for predicting future forecasts. This a necessary measure which creates financial forecasts based on historical data (Appendix 1), linear regression and Net Present value within this model formed essential tools for forecasting costs and revenues. The tools significantly enhanced the accuracy of financial forecasts as well as facilitating efficient in decision making on the firms budgetary process (Wijnhoven .n.d). Furthermore, Glazing and conservatory ads were tested to ascertain correlation based forecast output measures (see Appendix 2a and 2b). According to output results there is a positive correlation between both Ads and sales revenue returns. This implies that higher ad levels seem to be associated with increased sales revenue. Similarly, regression approach was applied on the data to confirm the linear regression assessment as well as aiding in generation forecasts. The visual graphical representation (Appendix 4) confirms this initial assessment. In order to determine the strength of relationships and extend of forecasts from the two measured, advertising strategies on sales revenue, it was necessary to determine the connection between the two variables. This is represented by the F values or coefficient of determination from regression analysis. The F value from NPV measures the degree of how each variable is associated with sales revenue and population in relation to the firm’s activity and costs which for the paradigm for decision making strategies aimed at improving management costs in the organization (see Appendix 3) (Wijnhoven .n.d). Conclusion This paper demonstrates that DSS in linear regression and Net present value model validates data bringing up new capabilities for management to make decisions more efficiently. Application of these tools assists the decision maker in integration of measured data and model results for financial forecasts, activating planning, costs determination and forging strategies which increase revenue returns. Thus DSS is a success avenue for organizational management quest to find solutions to scenario problems currently prevalent due to the nature of competition as a result of globalization. However, hybridization of DSS can assist management in preparation and mapping out business risks and handling of scenario problems better (Wijnhoven .n.d). Bibliography Knoo, B & Forgionne, G. n.d. Management Information Systems: Web-Based Decision Making Support Systems. Retrieved January 1, 2010 from ? Wijnhoven, F.n.d. Model Management by Means of Computer-based Information Systems in managerial contexts. Retrieved January 1, 2010 from Appendices Appendix 1 shows Historical data and forecasted yearly data Historical Data Year Pop'n D Glazing Adv Population Conservatory Adv Total Adv D Glazing Slaes Conserv'ry Sales Total Sales Cogs Tax Gross Profit Net Income 2003 704 18 704 22 40 £1,545 £1,543 £3,088 1515 413.856 1532.8 1119 2004 735 35 735 43 78 £1,646 £2,100 £3,746 1778 510.192 1889.6 1379 2005 775 50 775 78 128 £1,931 £1,840 £3,771 1788 500.742 1854.6 1354 2006 816 50 816 70 120 £1,979 £2,224 £4,203 1961 572.886 2121.8 1549 2007 882 70 882 85 155 £2,042 £1,975 £4,017 1887 533.304 1975.2 1442 2008 948 80 948 90 170 £2,310 £2,223 £4,533 2093 612.846 2269.8 1657 Forecasting Data 2009 979 94 979 94 187 £2,418 £2,255 £4,673 2149 631 2337 1706 2010 1028 103 1028 103 206 £2,543 £2,336 £4,879 2232 659 2442 1782 2011 1076 113 1076 113 226 £2,675 £2,419 £5,094 2317 688 2550 1861 2012 1125 124 1125 124 249 £2,814 £2,504 £5,318 2407 719 2662 1943 2013 1174 137 1174 137 274 £2,961 £2,593 £5,554 2502 750 2779 2028 2014 1222 151 1222 151 301 £3,117 £2,685 £5,801 2601 783 2900 2117 2015 1271 166 1271 166 331 £3,282 £2,780 £6,062 2705 817 3026 2209 2016 1319 182 1319 182 364 £3,458 £2,879 £6,337 2815 853 3158 2305 2017 1368 200 1368 200 401 £3,646 £2,982 £6,628 2931 890 3296 2406 2018 1416 220 1416 220 441 £3,847 £3,089 £6,936 3054 929 3441 2512 Appendix 2a shows summary out for yearly data spending on Advertisements SUMMARY OUTPUT Regression Statistics Multiple R 0.970954868 R Square 0.942753356 Adjusted R Square 0.904588926 Standard Error 85.77257623 Observations 6 ANOVA   df SS MS F Significance F Regression 2 363468.0288 181734.0144 24.70240916 0.013696987 Residual 3 22070.8045 7356.934833 Total 5 385538.8333         Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 564.1887976 1087.466357 0.518810347 0.639734526 -2896.614491 4024.992086 -2896.614491 4024.992086 Pop'n 1.220516148 1.776700098 0.686956763 0.541448813 -4.433736514 6.87476881 -4.433736514 6.87476881 D Glazing Adv 7.050028835 7.244088332 0.973211329 0.402228263 -16.00389331 30.10395098 -16.00389331 30.10395098 Appendix 2b. Shows regression correlation of the two variables SUMMARY OUTPUT Regression Statistics Multiple R 1 R Square 1 Adjusted R Square 1 Standard Error 7.3655E-14 Observations 10 ANOVA   df SS MS F Significance F Regression 1 194621.2043 194621.2043 3.58745E+31 6.762E-124 Residual 8 4.34004E-26 5.42505E-27 Total 9 194621.2043         Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 639.33 9.61199E-14 6.65138E+15 2.9236E-124 639.33 639.33 639.33 639.33 X Variable 1 48.57 8.10914E-15 5.98954E+15 6.762E-124 48.57 48.57 48.57 48.57 D Glazing Advertising% 0.5 Conservatory Advertising % 0.5 Appendix 3 shows Net Present value (NPV) against revenue stream. Discount rate 0.04 Year 0 1 2 3 4 5 6 7 8 9 10 Income stream -7000 £1,717 £1,794 £1,874 £1,957 £2,043 £2,133 £2,226 £2,323 £2,425 £2,532 Present value -7000 1650.99 1659.09 1666.40 1673.12 1679.40 1685.43 1691.37 1697.38 1703.63 1710.26                         Net Present Value 9817.08 981.71 Purchase Price 7000 (thousands)   1 2 3 4 5 6 7 8 9 10 int at bank 0.03                   bal 7000000 7210000.00 7426300.00 7649089.00 7878561.67 8114918.52 8358366.08 8609117.06 8867390.57 9133412.29 C/f 7210000 7426300 7649089 7878561.67 8114918.52 8358366.076 8609117.058 8867390.57 9133412.287 9407414.655 need name 2407414.655 need name 1791342.596 need name 981707.5292 need name 809635.0665 Appendix 4 shows graphical representation of regression analysis Read More