Drp_Case

Advanced Supply Chain Management Case Study Improving DRP effectiveness in ERP environment |Submitted To |Submitted By | |Prof. Mukundan |Joju Johny | | |Roll No 17 | | |Operations Management | | |MMS ( 2nd Year ) | | |BESIMSR | Introduction Competition has become intense in the household insecticides market due to a number of new players who have entered. Thus there is an urgent need for SPL reducing costs as customers now have a choice. This is where Distribution Requirement Planning (DRP) becomes important. If implemented successfully, DRP can enable planners to better anticipate future requirements of the depots, closely match the customers demand, more effectively deploy inventory to meet customer service requirements and adjust to the changing market. SPL has implemented ERP package. DRP is one of its functions of the distribution module of the ERP package. In spite of the distribution module in place DRP was not being utilized. Thus a study was initiated to quantify the benefits that SPL can derive from DRP so that the user can gain confidence in the DRP. It was also to review DRP process in an ERP environment to identify measures to improve its efficiency. SLP – Company Profile 1. SPL is the largest household insecticide player in the country and the world’s largest producer of mosquito repellent mats. 2. SPL is now in the growth phase. Its three major bands SP1, SP2, SP3 command almost 70 percent share of the mosquito repellent mats market, which is the largest category of household insecticides in India. 3. SPL’s market and product profile: • In the household insecticide market 80% is for the mosquito repellent products category while 20 % is for the insect killers. • The mosquito repellent category is subdivided as: ← 65 % - mats ( Growth rate : 10 % ) ← 20 % - coils ( Growth rate : 35 % ) ← 10 % - liquid vaporizers ( Growth rate : 50 % ) ← 5 % - aerosols ( Growth rate : 50 % ) 4. SPL’s logistics network: • SPL has 4 manufacturing sites at Pondichery, Goa, Silvassa and Nashik. • It has 4 regional offices in metros – Delhi. Mumbai, Chennai, Calcutta and is headed by a branch manager, who in turn reports to the national logistics manager (NLM). Each regional office has CFA’s under its jurisdiction. • Some products move from the factory to the depots and then to the further satellite depots while fast moving goods are directly dispatched to the depots. 5. Distribution Planning: • In SPL, distribution mainly constitutes of two types of materials movements, either from the factory to the depots directly or from the factory to the hub centers, from which it is then sent to the satellite depots. • The distribution planning procedure for these two movements of material are slightly different as the supply source for the forecast items is the factory in case of single items, while in case of two tier distribution it is a regional warehouse which acts as a supply source of CFA’s. Problem Definition 1) SPL has implemented the distribution module of ERP at a cost of Rs 3 crore. All CFAs have distribution module of ERP. Weekly sales and stocks arrive from CFAs electronically through emails. 2) Inspite of that, the DRP functions were not being used by the logistics department. 3) ERP has provided common bases for communicating information. But, only if DRP functionality is used, the benefits of having a common platform for information exchange can be more successful. 4) Thus the efficacy of the current manual dispatch plan was calculated by examining a few ratios like inventory turns, transportation, cost and interests cost from the balance sheet and the profit and loss account statement for the FY 1995 – 96 and 1996- 97. 5) These ratios clearly showed that the manual distribution plan was not the best plan. The ratios had marginally improved over the FY 1996 – 97. Transportation cost had increased and the inventory turnover was poor. 6) All these showed that there was scope for improvement and it was believed that DRP is one such tool which can improve this. It was believed that DRP is one such tool which can improve financial performance of SPL at no cost. 7) As SPL had already spent Rs 3 crore on implementing the distribution module, it was possible to improve SPL’s bottom line at no extra cost. |Ratio |1995 – 96 |1996 – 97 | |Inventory turnover |5.0 |5.9 | |Forward Cover |73.7 |62.9 | |Transportation Cost / Sales |1.0 % |1.8 % | |Interest Cost / Total Cost |2.6 % |2.5 % | Objectives 1) The reason for process users not utilizing DRP was the lack of confidence by the top management as well as the users. 2) Thus the project was initiated • to draw the top management attention towards the need for using DRP by showing its impact on the bottom line of the organization • to develop confidence in the users about the DRP functionality by demonstrating how it could improve the performance of the logistics department 3) To achieve these objectives the study was divided into 2 phases namely Phase I and Phase II Phase I Objectives: a. Develop a framework for benefit quantification. b. Simulating the distribution of SPL using DRP. c. Comparing the simulated distribution plans with actual distribution plans. d. Projecting the benefits and its impact on corporate profitability. Phase II Objectives: a. Identifying issues in current DRP practices. b. Reviewing inputs to DRP c. Analyzing the problem areas d. Recommending the improvements. Proposed Comprehensive Model for DRP 1. The job of DRP is to manage the flow of materials from supply sites to demand sites. 2. According to Martin, this was accomplished in three distinct phases i. Input Phase: DRP receives inputs like forecast, customer orders, inventory records and planning parameters for each SKU. ii. DRP process phase: After all the inputs are received, DRP generates a time phased model of resource requirements to support logistics strategy. iii. Output Phase: The DRP generates the planned order dictated by the item order policies and align the order by due date with demand which is known as planned order releases. In addition, it also generates action messages, summary and detailed reports. 3. The DRP process as described by Martin focuses on linkages of DRP outputs with manufacturing process. Kimball (1995) has also discussed DRP linkages with master scheduling. The model proposed by Ross (1996) provides a good insight about the internal mechanism of DRP and does not bring out the input and output variables strongly. 4. A good distribution model is the one which provides both internal mechanisms of the process and its linkages with the other processes of the organization. The proposed model of DRP is one such model. Measure for Quantification 1. A meaningful quantification would be one that shows an impact on the bottom line of the organization. Return of Investment is one such measure, which takes into consideration not only the impact on the bottom line but also on the cash flows and asset utilization. 2. Logistics department is the one whose impact could be seen in all these three financial aspects of the business and ROI could prove to be the right platform to project the benefits of DRP. 3. DRP can help in improving sales revenue by placing inventory at the right location thereby preventing lost sales. However, in SPL the practice is such that in order to prevent lost sales, inter CFA transfer are done which leads to additional cost of freight, loading and unloading. Thus DRP may not result in the improvement of sales revenue per se but improve profitability by reducing costs. 4. Transport cost consists of two cost components: primary dispatch TC (from factory to depot) and inter-CFA TC (from depot to depot). DRP has the potential to improve total transport capacity utilization. So any improvements in these will reflect in total TC. Simulation Exercise 1. In order to quantify the benefits of DRP, a simulation exercise was carried to prove that if DRP was used instead of the current manual system, then benefits like reduction in inventory, reduced transportation costs could be achieved. 2. The exercise was carried out for a fast moving SKU, SK1 under the following assumptions: • Weekly forecast is assumed to be available for eight weeks on rolling basis. • Movement of goods on routes not identified in the bill of distribution will be considered as inter CFA transfers and therefore avoidable costs. • The existing distribution networks viz direct dispatch from factory to depots is considered. 3. Simulation Parameter: The simulations used for the study are: a. Fixed Order Quantity (truckload). b. Planning horizon ( raw material procurement to receipt by CFAs – eight weeks ) c. Lead times as per contract agreement. 4. Data Requirement: The data required for the simulation exercise is as follows: a. Opening stocks and goods in transit b. Weekly sales and value of stocks. c. Actual dispatches from factory to CFAs. 5. Analysis Parameter: Number of dispatches a. Weekly inventory position. b. Avoidable dispatches – Inter CFA Impact on Corporate Profitability 1. Reduction in cost: When the DRP system was implemented it was found that there was a savings of 31.68 Lakhs as compared to the earlier records during which the DRP system was not used. This annual result was from the use of DRP for a single SKU. 2. Reduction in assets: By using DRP, the average weekly inventory had reduced in each region. Since inventory in the balance sheet is a snapshot picture, the average weekly inventory can be taken as it is for showing reduction in inventory. 3. Impact on ROI: It was found that, there was a 1.2 % increase in the ROI for SPL from the use of DRP module of the ERP system. This is a significant improvement considering the fact that this improvement is only from one SKU i.e. a major one. DRP Review 1. Further to see the impact of forecast accuracy on performance of DRP, a similar situation exercise was carried out. 2. It was found that with 100 percent accuracy DRP can further bring down the transportation and inventory carrying costs by 15 %. 3. Generalizing the conclusion it can be stated as: the correctness of the system is entirely dependent upon the correctness of parameters fed into the system. 4. Crawford (1993) has recognized certain inadequacy in the conventional DRP. Therefore, there is a need to review DRP to improve its effectiveness. 5. The first step in reviewing the DRP is to check the validity of data and planning parameter. 6. Stein (1996) proposed that for an information system to resemble reality, it must be dynamic in nature. 7. However, data are instead stored as static and recalled from the database without the consideration of the changes which may have occurred or the physical laws which govern how it should be used. 8. Therefore, to improve effectiveness, those inputs that are dynamic, need to be identified and treated accordingly. 9. Using this methodology suggested by Ross (1996), it was found that the dynamic inputs to DRP are Safety Stock (SS), Lead time (LT), Bill of Distribution (BOD) and ordering Policy (OP). These parameters were analyzed in detail and the summary of the analysis is presented next. Safety Stock The safety stock was calculated using the analytical approach as suggested by Plossl (1986) with the following assumptions: 1. The demand is normally distributed. 2. Average of sales is close to average of the forecast. The analysis will be done by taking average of the sales as the forecast. So all the other requirements of normal distribution will be satisfied. 3. Desired service level is 95 percent. A closer look at sales trend of SPL reveals that the year could be split into smaller periods wherein demand in each period is more uniform than considering the full year as one period. When the review period was reduced from 12 months to 3 months, the safety stock levels can be brought down drastically. Thus DRP system is being made dynamic to a certain extent. Lead Time Variation 1. For the actual lead times, the mean and standard deviation were calculated. The lead time is assumed to be normally distributed. 2. The lead times were calculated and it was concluded that depots far from factory receive material late by 2 days on an average. The lead times currently fed into the system were as per contract agreement. They need to be changed to the mean value. Ordering Policy (OP) 1. SK being a fast moving SKU is sent directly to depots in truckloads. So the ordering policy can be said to be Fixed Order Quantity (FOQ) with an ordering quantity as a truckload. 2. Actually by shipping truckload quantities rather than less than truckload quantities, a company may experience lower transportation rates only as long as any expenses exceed the savings in transportation costs. This needs to be examined. 3. This trade off needs to be worked out depending upon the ordering/ shipping frequency. If shipping were frequent, it would be costlier, but less than truckload will be dispatched as frequency increases. Consequently, inventory carrying cost decreases. 4. Four cases have been analyzed viz, one dispatch every week / every 2 weeks / every 3 weeks / 4 weeks. Calculations of transportation were done for each quarter. 5. From the results, it was seen that for the first quarter, it was economical to make dispatch every week to Calcutta from factory. However, for the rest of the year, it was economical to make dispatches once every second week. Ordering Quantity 1. It was found that the ordering policy which was followed at SPL for SK was not an optimum policy. 2. Periodic Order Quantity (POQ) was the ordering policy where a time period is specified instead of quantity. This ordering policy could be used along with the trade off results. 3. From the above analysis, it was clear that the best ordering policy for SPL would be POQ, where period should be defined based upon the optimum shipment frequency. The Proposed Semi Dynamic Model 1. Based on the parameters mapping results, it was found that parameters should be changed every quarter. 2. To incorporate this dynamic feature in the comprehensive model proposed by us, one has to modify the model. 3. The model shown below is renamed as Semi Dynamic Model for DRP process (SDMD) generalizes the findings for SPL, where the planning parameter fed to DRP are function of sales trend and/or time. Recommendations 1. Planning process: The distribution planning should be done centrally by taking DRP run at every HO every week. 2. Using DRP to generate MPS to synchronize operations: The integration of DRP with MPS gives a single, continuous seamless system that uses one set of logic across the operations. DRP integrates logistics, manufacturing and purchasing thereby contributing to ability of users to effectively control ongoing activities by comparing plans against operating budgets. 3. Revising DRP parameters every quarterly: SPL’s product exhibits seasonality and if parameters are not revised quarterly, the parameters may not be optimum for a particular service level. 4. Using safety stocks for demand uncertainties, calculated at 95 % service level. 5. Using safety time for lead time uncertainties, calculate at 95 % service level. 6. Following periodic order quantity (POQ) dispatch policy, where period shall be calculated for optimum total cost from a trade off analysis between transportation cost and inventory holding cost. 7. The time bucket of one week should be uniformly applied: Since the DRP is run on a weekly basis; the factories dispatch plan should be reviewed on a weekly basis. 8. Dispatch variance report should be generated by factory, giving reasons for any dispatches not done as per the dispatch plan. So that the distribution planner comes to know the constraints under which the factory operates. The planner can then take these constraints into account while making future plans. 9. Provide the DRP summary information to depots so that even the branch managers have the visibility of the system and can suggest changes in the forecasts or lead times which according to them are not correct. These are the people who have tremendous knowledge about products and customers.