Operation Management - Delta Plastics, Great Northwest Outdoor Company - Essay Example

Table of Contents Portfolio Exercise Delta Plastics Inc Portfolio Exercise 2: Great Northwest Outdoor Company7 Portfolio Exercise 3: Professional Video Management.13 Portfolio Exercise 4: Newmarket International Manufacturing .16 Portfolio Exercise 1 Delta Plastics Inc. In the area of manufacturing, not only is the reliability of the process important but also the consistency. It is therefore necessary to have some measure on how controlled a process is. Statistical process control provides a good framework which can be used in analyzing processes. Among the different tools available are the -charts, R-charts, p-charts, and c-charts (Reid and Sanders). Both the -charts and R-charts are used to determine whether the output of the process fall within a certain range. These measures may focus specific features such as the net weight of the product. Alternatively they may also analyze the process as a whole such as the average output (Heizer and Render). However, what these two charts often do not measure is the defect rate. For this reason, p-charts and c-charts are present to do such analysis. The difference between the two is that the former can be used to describe the proportion of the population which is defective (Reid and Sanders). On the other hand, c-charts are used to analyze the raw quantity of defects which occur. In the case presented, only the number of defects is readily available for analysis. It is therefore necessary to use c-charts in analyzing data. There can be two approaches to analyzing the said data. To view the process as a whole, the total daily defect count can be computed. Using the daily defect count, the mean may be obtained. Subsequently, the upper and lower control limits for the c-chart may be constructed and analyzed. Tables 1 and 2 present the daily defect data for the standard material and the super plastic material. Table 1. Total Defect Analysis for Standard Material Monday Tuesday Wednesday Thursday Friday Week 1 11 8 12 12 12 Week 2 12 8 6 11 9 Week 3 12 7 8 6 13 Week 4 11 8 6 12 9 Table 2. Total Defect Analysis for New Material Monday Tuesday Wednesday Thursday Friday Week 1 11 9 11 8 12 Week 2 11 10 9 13 11 Week 3 10 14 17 12 10 Week 4 15 16 12 15 15 Using this material, the lower and upper control limits may be determined as follows: Knowing these quantities, the control charts shown in Figure 1 and 2 may be constructed. Figure 1. Control Chart for the Defects in the Standard Material Figure 2. Control Chart for the Defects in the New Material These two control charts reveal some interesting features of the two processes. The process for manufacturing the standard material shows a stable operation. The points in the control chart are centered on the mean value and fluctuate above and below the central tendency. Also, no spikes can be seen in the chart. It can therefore be said that the original process is in control. The second chart shows a trend quite similar to the original process. As with the first chart, there are no spikes above or below the control limits. Also, the data fluctuates as with any normal process. However, it should be noted that near the start of the production process, the points are generally below the control limit. As each successive day in the production process passes, the total number of defects appears to be increasing. In fact, at the second half of the data, the points tend to be above the mean. The control chart in Figure 2 therefore reveals a drift towards the upper control limit. While this process is currently in control, it may eventually shift above the upper control limit. It therefore warrants further investigation (Heizer and Render). To isolate the nature of the defect, it becomes helpful to analyze each type of defect. That is, a control chart may be constructed for each particular type. The said defect counts are shown in Tables 3 and 4. Table 3. Individual Defect Count for the Standard Material Standard Uneven Crack Scratch Bubbles Thickness Week 1 M 1 2 3 4 1 T 2 3 1 2 0 W 2 2 2 2 4 Th 3 3 3 3 0 F 2 0 4 4 2 Week 2 M 3 3 2 4 0 T 1 2 1 3 1 W 1 2 0 2 1 Th 2 1 2 4 2 F 0 3 3 3 0 Week 3 M 2 3 3 3 1 T 2 2 1 1 1 W 1 2 0 2 3 Th 3 0 1 2 0 F 2 2 3 4 2 Week 4 M 3 1 3 2 2 T 1 2 1 3 1 W 1 2 0 2 1 Th 2 3 1 4 2 F 2 3 3 1 0 Type Total 36 41 37 55 24 Table 4. Individual Defect Count for the New Material Super Uneven Crack Scratch Bubbles Thickness Week 1 M 2 6 0 2 1 T 2 6 1 0 0 W 3 4 0 2 2 Th 2 3 1 1 1 F 0 7 0 3 2 Week 2 M 3 4 0 4 0 T 1 4 1 3 1 W 1 4 0 2 2 Th 2 3 2 4 2 F 3 3 2 3 0 Week 3 M 1 4 0 4 1 T 2 6 1 5 0 W 2 4 2 5 4 Th 2 4 1 5 0 F 3 3 0 3 1 Week 4 M 3 5 1 6 0 T 2 7 1 5 1 W 1 6 0 4 1 Th 2 3 2 6 2 F 0 7 3 5 0 Type Total 37 93 18 72 21 Again, the upper and lower control limits may be determined from the mean values of each defect. The collective upper and control limits are shown in Table 5. Table 5. Upper and Lower Control Limits for Each Defect Type Uneven Crack Scratch Bubbles Thickness Daily Total Standard Material Upper Limit 5.824922 6.345346 5.930441 7.724937 4.486335 18.969335 Standard Material Lower Limit -2.23 0 -2.25 0 -2.23 0 -2.23 0 -2.09 0 0.3306653 Super Plastic Upper Limit 5.930441 11.11916 3.74605 9.2921 4.124085 22.463933 Super Plastic Lower Limit -2.23 0 -1.82 0 -1.95 0 -2.09 0 -2.02 0 1.636067 It is then necessary to obtain the control charts for each defect type for each of the two materials. These charts are comparatively presented in Figure 3. Using the same manner of analysis as the previous case, it can be seen that the occurrence of uneven edges, cracks, scratches and thickness variations are in control for both materials. These control charts swing consistently about the mean value. Also, no consistent occurrences of these defects occur above or below the mean line. However, by looking at the defect counts for air bubble defects in the new material, it can be seen from the control chart that a trend similar to the total defect count is occurring. This indicates a possible fault in the machinery or the process itself. Considering that the same defect is in control for the old material, this limits the root of the problem to new equipment acquired for the super plastic or to processes that differ significantly between the two materials. It is these areas that should be the concern of the company and should be properly investigated to reveal any issues regarding the new material. Portfolio Exercise 2 Great Northwest Outdoor Company In both manufacturing and sales organizations the forecasting of demand plays a crucial role in the management of organizations (Heizer and Render). Improper forecasting could lead to great losses on the part of the company. For instance, if the forecast is much greater than the actual demand, overstocking may occur and excess materials would be wasted. This is even more so in some cases such as the handling of perishable items. On the other hand, the underestimation of demand would result to an insufficient supply of items which would mean lost sales opportunities and possibly customer dissatisfaction. There are several means to predict the demand for a given time period (Reid and Sanders). The method which should be adopted is dependent on many factors including the actual nature of the item. Some items can be sold well during certain seasons and others are easily in demand all throughout the year. Also, some items are consistently sold for successive time periods while others will exhibit and rise or decline in demand through time. Of the many methods, one of the commonly adapted methods is the use of exponential smoothing. This method takes into consideration the previous forecast and actual demand to generate a present prediction. Generalizing this technique, the current forecast is taken as the previous forecast with a corrective term equal to the error of the previous value to predict the actual demand. This method is often a very good forecasting tool but it is limited to non-seasonal data. The case of Great Northwest Outdoor Company uses a special form of exponential smoothing. Rather than analyzing the demand for the year as a whole, it takes into consideration seasonal demand by taking each quarter separately. Four unique estimates are then produced for each year. This results to a good estimate on the demand for each season as these generally show a stable pattern. In Figure 1, the demands for each quarter from the years 2000 to 2004 are shown. In this chart, it can be seen that in general, the demand for each season is rising. Figure 1. Quarterly Demand for the Years 2000-2004 However, a major limitation of the use of exponential smoothing is the inability to cope with rising or falling trends of the demand. In Table 1, the forecasted and actual demands are presented and analyzed with the mean absolute deviation (MAD). These deviations are also graphically depicted in Figure 2. It can be seen that for each successive year, the error in forecasting actually increases. It is therefore necessary to consider some form of trending together with seasonal adjustments. Table 1. Exponentially Smoothed Forecast vs. Actual Demand for 2000-2005 and their Respective Deviations Qtr 1999 2000 2001 2002 2003 2004 2005 1st 18.5 18.5 18.53 18.401 19.6007 20.68049 21.82634 2nd 23.4 23.4 23.43 23.811 25.3077 25.99539 27.49677 3rd 20.1 20.1 20.19 19.983 20.2881 21.52167 22.17517 4th 41.5 41.5 41.62 43.024 43.7668 44.76676 47.17673 (a) Exponentially Smoothed Forecast with Qtr 2000 2001 2002 2003 2004 1st 18.6 18.1 22.4 23.2 24.5 2nd 23.5 24.7 28.8 27.6 31 3rd 20.4 19.5 21 24.4 23.7 4th 41.9 46.3 45.5 47.1 52.8 (b) Actual Demand Qtr 2000 2001 2002 2003 2004 MAD 1st 0.1 0.43 3.999 3.5993 3.81951 2.389562 2nd 0.1 1.27 4.989 2.2923 5.00461 2.731182 3rd 0.3 0.69 1.017 4.1119 2.17833 1.659446 4th 0.4 4.68 2.476 3.3332 8.03324 3.784488 (c) Deviation Figure 2. Absolute Deviations for the Years 2000-2004 for Exponential Smoothing One alternative solution is the use of seasonally adjusted forecasting. This method assumes that the demand for each season follows a fixed percentage of the total yearly demand. A seasonal index is derived for each season. This index is then multiplied the average demand per season within the year to determine the forecast. However, it is apparent that this method requires some forecast of the yearly demand. This is often obtained through a linear estimate of the trend. Using linear regression, a trend line is constructed and can be used to predict future demands. For the Great Northwest Outdoor Company, the trend is forecasted as: The seasonal indices to be used for forecasting are derived in Table 2. Table 2. Seasonal indices for Great Northwest Outdoor Company Qtr Index 1st 0.730256 2nd 0.927179 3rd 0.745299 4th 1.597265 With this information, the forecast for each quarter from 2000 to 2005 can then be computed as presented in Table 3. Also, the deviations from the actual demands can be seen in the same table as well as in Figure 3. Table 3. Seasonally Adjusted Forecast vs. Actual Demand for 2000-2005 and their Respective Deviations Qtr Index 2000 2001 2002 2003 2004 2005 Total 96.33 103.22 110.11 117 123.89 130.78 1st 0.7303 17.58745 18.84539 20.10333 21.36128 22.61922 23.87716 2nd 0.9272 22.32929 23.9264 25.5235 27.1206 28.7177 30.3148 3rd 0.7453 17.94869 19.23247 20.51625 21.80003 23.0838 24.36758 4th 1.5973 38.46698 41.21833 43.96968 46.72103 49.47237 52.22372 (a) Seasonally Adjusted Forecast Qtr 2000 2001 2002 2003 2004 1st 18.6 18.1 22.4 23.2 24.5 2nd 23.5 24.7 28.8 27.6 31 3rd 20.4 19.5 21 24.4 23.7 4th 41.9 46.3 45.5 47.1 52.8 (b) Actual Demand Qtr 2000 2001 2002 2003 2004 MAD 1st 1.01255 0.745391 2.296667 1.838725 1.880783 1.554823 2nd 1.170706 0.773604 3.276502 0.4794 2.282298 1.596502 3rd 2.451313 0.267534 0.483754 2.599975 0.616196 1.283754 4th 3.433023 5.081674 1.530324 0.378975 3.327626 2.750324 (c) Deviation Figure 3. Absolute Deviations for the years 2000-2004 for Seasonally Adjusted Forecasting As compared to exponential smoothing, this alternative forecasting approach minimizes the forecasting error. This is due to the fact that the trend is considered in the estimate. However, a major limitation of seasonally adjusted forecasting is that it uses a fixed trend. This may be remedied by using more complex trend equations to minimize the errors. What this method does not consider are changes in seasonal demand. If for the past few years, the holiday season's sales have been declining, this forecasting method cannot take the said seasonal demand adjustments into consideration. A third option for the company is one that takes the advantages of each of the two previous methods. By incorporating trend into exponential smoothing, a sounder estimate is obtained. Both a trend and a forecast are computed. For the purposes of calculation, we assume that the initial trend is the failure of the previous model to estimate the changes in annual demand. That is, the mean absolute deviation becomes the initial estimate for the trend of each respective season. The value for in this new forecasting model is chosen to minimize the new MAD. From testing different values, a value of = 0.1 is obtained for the current smoothing factor . The comparison of the predicted values and the actual demand are shown in Table 4. The deviations are also presented in the said table and graphically illustrated in Figure 4. Table 4. Trend-Adjusted Exponentially Smoothed Forecast vs. Actual Demand for 2000-2005 and their Respective Deviations Qtr 1999 2000 2001 2002 2003 2004 2005 1st 18.5 18.5 20.3172 21.42631 23.08367 24.47541 25.74857 2nd 23.4 23.4 25.4734 27.18045 29.2997 30.55166 32.11925 3rd 20.1 20.1 21.41955 22.10775 22.80331 24.03974 24.81908 4th 41.5 41.5 44.43838 47.55178 49.59295 51.21685 53.55102 (a) Trend-Adjusted Exponentially Smoothed Forecast Qtr 2000 2001 2002 2003 2004 1st 18.6 18.1 22.4 23.2 24.5 2nd 23.5 24.7 28.8 27.6 31 3rd 20.4 19.5 21 24.4 23.7 4th 41.9 46.3 45.5 47.1 52.8 (b) Actual Demand Qtr 2000 2001 2002 2003 2004 MAD 1st 0.1 2.2172 0.97369 0.116331 0.024587 0.686361 2nd 0.1 0.7734 1.619555 1.699695 0.448344 0.928199 3rd 0.3 1.91955 1.107746 1.596688 0.339739 1.052744 4th 0.4 1.861625 2.051778 2.492947 1.583147 1.677899 (c) Deviation Figure 4. Absolute Deviations for the Years 2000-2004 for Trend-adjusted Exponential Smoothing It can be seen that the deviations are in fact minimized as compared to the previous two methods. This third alternative holds several advantages. For one, since the trend is also exponentially smoothed, it is capable of modeling more complex changes in trend. Also, the forecasting itself is already robust in the sense that it can also work with complex patterns. Finally, the forecast of this method does not revolve around the assumption that the demand for each season is constant relative to the annual demand. This last feature offers the greatest advantage over the seasonally adjusted forecast. It is therefore recommended that Great Northwest Outdoor Company should adopt this modification of its existing forecasting technique. Portfolio Exercise 3 Professional Video Management Resource management plays a vital role in most organizations. In the area of manufacturing and production, these resources include the raw materials needed to create their output. The proper management of these materials can actually pay a role in reducing the overall costs of the company. With raw materials, it is usually necessary to have some form of inventory. This is particularly true in the case wherein every order is associated with a setup cost. It is therefore necessary to reduce the amount of orders initiated within a given period of time. However, this is not fully realizable due to that fact that holding inventories also have associated costs. Each unit of an item in the inventory occupies an amount of space which is effectively an expense. Furthermore, the handling of these materials in inventory would also contribute to additional costs on the part of the organization. When dealing with material inventories, it is therefore necessary to balance the ordering costs and the carrying costs (Heizer and Render). One of the basic models used to optimize the ordering process is the use of Economic Order Quantities (EOQ). This model assumes that the demand follows a particular trend. Also, upon ordering, a fixed lead-time is given before the actual receipt of the items. It is also important to note that all items are received in one batch. To minimize the costs, the EOQ model ensures that deliveries arrive just at the time when the total stock is at a zero level. This effectively controls the holding cost of the items. An optimal quantity or EOQ is then used as the size of each batch such that the cost of ordering and the cost of carrying are balanced. This means that the number of times an order is placed for a given year is adjusted by the size of the batches. In the given case, two alternative options for the supply of materials are given. Steve Goodman is left with the decision of choosing between the companies Toshiki and Kony for the supply of the basic videotape systems. Unlike the traditional EOQ model, the case involves a more realistic factor - discounts. An alternative is to use the quantity discount model to effectively choose the best order quantity. Comparing the two companies, the primary edge of Toshiki is that it offers a lower price for very large batches as compared to Kony. However, the pricing scheme of Kony permits the purchase of fewer materials at the same price as Toshiki. Another advantage Kony has over its competitor is its close proximity to Steve's operation which means that the ordering cost is reduced. For more quantitative data, the optimal order quantities for each quantity range for Toshiki and Kony are derived as follows: Table 1. Monthly Demand and Average Month Demand June 7970 July 8070 August 7950 September 8010 Average Demand 8000 Table 2. Demand for Different Time Periods Monthly Daily Weekly Yearly Whole Unit 8000 400 2000 96000 Video Tape System 16000 800 4000 192000 Table 3. Economic Order Quantity for Toshiki and Kony Q* Toshiki 0-2000 2000-8000 8000-20000 20000+ p = 250 p = 230 p = 220 p = 210 678.82251 707.72139 723.627227 740.65608 679 2000 8000 20000 Kony 0-1000 1000-5000 5000+ p = 240 p = 230 p = 220 461.880215 471.81426 482.418151 462 1000 5000 Taking these quantities into consideration, the alternative with the lowest annual cost is chosen as the best option. These quantities are computed below. Table 4. Total Annual Cost for Varying Order Quantities Total Cost Toshiki 0-2000 2000-8000 8000-20000 20000+ 48050911.7 44237640 42506160 40950864 Kony 0-1000 1000-5000 5000+ 46113255.4 44202180 42406536 It is therefore apparent from these comparisons that despite the high ordering cost of Toshiki, they still provide the best solution considering the fact that they offer lower prices at large quantities. However, another concern with the distance of Toshiki is the lead time. With the given data that it takes three whole months for a shipment from the said company to arrive, the reorder point is to be considered. To compare with Kony, the following computations are shown: Table 5. Reorder Points ROP Toshiki 48000 Kony 8000 Seeing as the reorder point is actually above the chosen EOQ for each company, it is impossible for the quantity in the inventory to reach such a level. It is therefore important that the orders are held at regular intervals significantly before the item is needed (Winston). Keeping this in mind, the actual reorder points can be obtained in the following solution: Table 6. Modified Reorder Points ROP/Q* new ROP Toshiki 2.4 8000 Kony 1.6 3000 With these new values, it can be seen that both companies can realistically be chosen as suppliers. As such, the choice of Toshiki as the main source remains. Finally, a last train of thought to consider would be to have different production approaches. One alternative is to individually sell the components for the video system. This move would reduce the need for inventories. The primary motivation for this shift is that a complete video system does not necessarily have to be assembled. Therefore, it is no longer necessary for a fixed proportion of parts to be obtained from suppliers and put together. Instead, each individual component may be purchased from the suppliers based on the demand and the optimal quantity. This effectively reduces the cost of inventories. Another approach would be enabling the control box to interface with other videotape systems while providing the option of using the ones from Toshiki or Kony. This decision would reduce the effective demand for the videotape system from Steve. Consequently, the optimal order quantity may shift to a lower bracket with a higher price. Ultimately, depending on the demand, it may lead to the choice of Kony as the best supplier as the 20,000+ may be unnecessary. Also, the reorder points would shift. If the daily demand multiplied to the lead time is still above the EOQ, the reorder point would shift depending on the number of order cycles in a year. However, if the said expression goes below the EOQ, it then becomes fixed at that value. Portfolio Exercise 4 Newmarket International Manufacturing (A) In an organization the amount of money spent on labor costs often sum up to a sizable amount. For this reason, it is important to keep track of the budget spent on labor and if possible minimizing it. Improper management of the workforce could mean unnecessary costs such as excessive labor. On the other hand, a deficiency in the number of workers in a manufacturing firm would also contribute to losses in the form of lost production. Since the workers form a critical part of the production process, it is necessary to ensure that they are properly managed. In Newmarket International Manufacturing (NIMCO), an interesting case is revealed. Processes are typically within the range of one of three manufacturing approaches - process-oriented, modular, or product-oriented (Heizer and Render). With these paradigms, the amount being produced is inversely proportional to the amount of customization that the process can accommodate. However, a special case of manufacturing is in mass-customization. This approach leads to large profits if properly executed. However, the nature of this type of processing does not allow for inventories which means that products should be delivered just as they are needed. This just-in-time approach may lead to difficulties on the part of the labor force of a company. In NIMCO's case, the past quarters was shown to be quite undesirable. While the brief does not give exact details, one possible cause of the poor performance is the inability of the labor force to cope with the changing demands. To resolve this problem, proper forecasting of demand must be made to ensure that the right amount of labor is available to complete the products in time to be sold. Three unique schemes were proposed to deal with the labor distribution. The first of these schemes was to use a level workforce. This means that the maximum demand is taken and the number of workers is chosen based on this maximum demand. This particular approach minimizes the costs of firing workers. Also, since a sufficient amount of labor is readily available, there is no need for employees to use overtime to deal with the demand. This naturally avoids the expensive rates at which personnel are compensated for overtime periods. In fact, the only direct costs in this scheme is the continuous payment of personnel despite non-working periods as well as the hiring of new employees should the existing labor force deem insufficient. In Table 1, a listing of the total number of work hours necessary for each week in the incoming quarter is presented. This data is used to compute for the number of employees needed based on the premise that each employee engages in 40 hours of work in a week. The necessary labor force for each week is presented in Table 2. Table 1. Work Hours in a Week Week Product A Product B Product C Work Hrs for A Work Hrs for B Work Hrs for C Total Work Hrs 14 3600 4000 2000 864 1520 580 2964 15 4000 4000 2500 960 1520 725 3205 16 4300 4000 2800 1032 1520 812 3364 17 4400 3800 3100 1056 1444 899 3399 18 4500 3800 3200 1080 1444 928 3452 19 4500 3800 3200 1080 1444 928 3452 20 4400 3600 3200 1056 1368 928 3352 21 4300 3600 3000 1032 1368 870 3270 22 4000 3600 3000 960 1368 870 3198 23 4000 3800 2800 960 1444 812 3216 24 3600 3800 2800 864 1444 812 3120 25 3200 3800 2600 768 1444 754 2966 26 3000 4000 2600 720 1520 754 2994 Table 2. Weekly Number of Workers Needed Week Employees Needed 87 14 74.1 75 48720 15 80.125 81 48720 16 84.1 85 48720 17 84.975 85 48720 18 86.3 87 48720 19 86.3 87 48720 20 83.8 84 48720 21 81.75 82 48720 22 79.95 80 48720 23 80.4 81 48720 24 78 78 48720 25 74.15 75 48720 26 74.85 75 48720 Total Labor Cost 633447 From the computed data, the maximum number of employees is 87 and using this first plan, it is therefore necessary to constantly hold the total number of employees for the quarter constant. However, it is clear that losses are incurred during most of the other weeks wherein a full workforce is not necessary. An alternative plan is to keep the existing workforce and to allow the employees to engage in overtime to cope up with larger demands. However, the disadvantage to this is that the extra hours spent by the workers in overtime have larger rates. At some point, the amount of extra pay to the current workforce may be equal to the losses incurred by using a level workforce; however, it can be seen in Table 3 that the current demand forecasts are below the said point. It is therefore more economical for the workers to simply use overtime to deal with extra demand. Table 3. Overtime Rate of Plan 2 Week Hrs/Employee Normal Excess Hrs Overtime Total 14 39.52 40 42000 0 0 42000 15 42.73333 43 42000 3 4725 46725 16 44.85333 45 42000 5 7875 49875 17 45.32 46 42000 6 9450 51450 18 46.02667 47 42000 7 11025 53025 19 46.02667 47 42000 7 11025 53025 20 44.69333 45 42000 5 7875 49875 21 43.6 44 42000 4 6300 48300 22 42.64 43 42000 3 4725 46725 23 42.88 43 42000 3 4725 46725 24 41.6 42 42000 2 3150 45150 25 39.54667 40 42000 0 0 42000 26 39.92 40 42000 0 0 42000 Total Labor Cost 616875 Another advantage of this approach is that it can deal to unexpected changes in demand. If the actual demand is lower than the forecasted demand, the company is able to save up more on overtime salaries. On the other hand, increased demand can be dealt with through the use of overtime. This is unlike the first proposed plan wherein a level workforce means even more losses when a dip in demand occurs. A third alternative to be considered is the dynamic hiring and firing of workers. This approach may seem quite expensive due to the hiring and firing costs of each worker. However, as compared to the cumulative overtime salaries of the previous plans, it can be determined that this approach is actually cheaper than the other alternatives. Table 4 shows that a minimum amount is spend on workforce adjustment. Table 4. Cost of Workforce Adjustment Week Employees Needed Hire Fire Salary Fees Total 14 74.1 75 0 0 42000 0 42000 15 80.125 81 6 0 45360 3000 48360 16 84.1 85 4 0 47600 2000 49600 17 84.975 85 0 0 47600 0 47600 18 86.3 87 2 0 48720 1000 49720 19 86.3 87 0 0 48720 0 48720 20 83.8 84 0 3 47040 2250 49290 21 81.75 82 0 2 45920 1500 47420 22 79.95 80 0 2 44800 1500 46300 23 80.4 81 1 0 45360 500 45860 24 78 78 0 3 43680 2250 45930 25 74.15 75 0 3 42000 2250 44250 26 74.85 75 0 0 42000 0 42000 Total Labor Cost 607050 While this may seem attractive, it is also important to consider how such an approach would deal with changing demands. Since the number of employees needed are already based on a forecasted demand, unexpected fluctuations in the actual demand would lead to problems. An increase would force the company to use overtime for their workers. Drop in demand would lead to losses with unutilized work hours. Considering the welfare of the customer, cost and the operation performance, the most stable plan is the use of overtime. This not only has a more manageable cost, it also adapts to changing customer demands. An additional advantage to this is that the existing workforce is retained. This means that the workforce retains its current level of experience delivering the best performance in terms of operation and customer support. Generally, the second plan remains the most flexible and practical of the three options. Bibliography Heizer, Jay and Barry Render. An Introduction to Operations Management. 8th. Pearson Education South Asia, 2007. Reid, Dan and Nada Sanders. Operations Management: An Integrated Approach. 2nd. Wiley & Sons, Inc., 2005. Winston, Wayne L. Operations Research: Applications and Algorithms. 4th. Cengage Learning, 2003. Read More