- Email: [email protected]
Calculating the Cost of Variances in the Supply Chain
Calculating the Cost of Variances in the Supply Chain Determining Supplier and Buyer Effect on Inventory Performance
Mary Margaret Weber The purpose of this work is to provide a framework that can be used to diagnose the causes of inventory-related variances in the buyer–seller relationship, and to analyze the costs to the channel associated with these variances. A multi-echelon approach is used to highlight variances caused by both supplier and buyer in increasing detail. The framework’s usefulness lies in its ability to separate variances caused by inadequate planning from those caused by poor performance. © 2000 Elsevier Science Inc. All rights reserved.
Address correspondence to Mary Margaret Weber, Emporia State University, School of Business, 1200 Commercial Street, Box 58, Emporia, KS 66801. Tel: 316-341-5315. E-mail: [email protected]
Industrial Marketing Management 29, 57–64 (2000) © 2000 Elsevier Science Inc. All rights reserved. 655 Avenue of the Americas, New York, NY 10010
INTRODUCTION The logistics practices of the past 15 years have made it clearly evident that in highly competitive domestic and international markets, the development of strong channel relationships is essential for the development and maintenance of competitive advantage. This trend toward strong channel relationships is thoroughly discussed in the logistics literature, for example [1–5]. Much of this push for stronger and long-term channel relationships focuses on the provision of superior customer service through better inventory management policies. Thus, new methods of evaluating channel performance that explicitly detail the impact of each channel member’s performance on the entire channel are needed [6, 7]. In particular, it has
0019-8501/00/$–see front matter PII S0019-8501(99)00112-1
Poor inventory planning affects other firms in a distribution channel. become increasingly important to eliminate variance from the channel relationship. In other words, increased competitive demands necessitate performance at high levels with decreasing acceptance of error. The purpose of this work is to provide a framework that can be used to diagnose the causes of inventoryrelated variances in the buyer-seller relationship, and to analyze the costs to the channel associated with these variances. A multi-echelon approach is used to highlight variances caused by both supplier and buyer in increasing detail. The framework’s usefulness lies in its ability to separate variances caused by inadequate planning from those caused by poor performance. Multi-echelon approaches were developed independently by Shank and Churchill [8] and Hulbert and Toy [9]. The Shank and Churchill model, later extended by Shank and Govindarajan [10], suggests that an analysis of variance in sales and marketing activities can benefit by adding increasing levels of complexity to enhance insight into the causes of variances. Each subsequent level of the analysis reduces the likelihood that large offsetting variances might be misinterpreted as a small, possibly insignificant, variance. A weakness of this work is that comparisons are made only between planned and actual performance, without an attempt to incorporate changes in the plan during the period being analyzed. Thus, this work allows no distinctions to be made between variances caused by inadequate performance and those caused by poor planning.
MARY MARGARET WEBER is Assistant Professor of Marketing at Emporia State University. She received her Ph.D. in Logistics from The Ohio State University and her MBA from Virginia Polytechnic Institute and State University. Her current research interests include the impact that information technology is having on the competitive structure of retailing and on the logistics practices throughout the marketing channel.
58
Hulbert and Toy’s work, later extended by Bentz and Lusch [11] and Lusch and Bentz [12], incorporates the idea of a revised plan into a multi-echelon approach. The use of a revised plan to evaluate forecasting accuracy is based on the broadly accepted concept of flexible budgeting, and it provides a means of recognizing the opportunity costs that arise from errors in forecasting [13–15]. Planning errors become apparent through comparison between the original and revised plans. While comparisons between actual and planned results can still be made to evaluate management performance, use of a revised plan eliminates the ability to hide any slack that was intentionally or unintentionally built into the original plan. The result is that a planning variance can be used to distinguish between the actual and forecasted operating environments. A performance variance distinguishes between actual performance and the performance expected, given the actual, as opposed to forecasted, environment. Weber [16] applied the multi-echelon approach to the evaluation of inventory-related variances in the supplier– buyer relationship. The model that was developed used actual, planned, and requested quantities of inventory to isolate causes of inventory surpluses or shortages. The model successfully distinguished between variances caused by inadequate supplier performance and those caused by errors in buyer planning. However, a major weakness of this framework is its use of placeholder values representing a combination of stockout quantity and time; it does not clearly elucidate costs associated with inventory-related variances. The framework proposed in this paper incorporates actual inventory valuations. Its use of both planning and performance variances helps to determine channel member responsibility for system variance. It also demonstrates how another channel member may compensate for planning or performance errors by one channel member and the cost of that compensation. The increasing levels of detail provide management cues for action in reducing overall system variance.
The variance model improves management’s ability to isolate and correct the causes of those variances. FRAMEWORK FOR VARIANCES IN TOTAL INVENTORY VALUATION Some of the same assumptions used by Weber [16] also apply to this framework. These include: 1)analysis of one product and one supplier–buyer relationship at a time; 2)the buyer–supplier relationship is such that buyer forecasts for the given time period are shared with the supplier; and 3)all variances are viewed as undesirable. The framework, itself, is shown in Figure 1. Positive
FIGURE 1.
variances represent inventory surpluses or opportunity costs associated with planning for excess inventory. Negative variances represent inventory shortages.
Variable Definitions Variables used in the framework are shown in Table 1. In addition, all variables are subscripted. Actual performance during the period is represented by a subscript “a.”
Fatal system variance.
59
The flexibility of suppliers in responding to shifting buyer needs can be determined. Thus, Xa indicates the actual number of units shipped/received. Planned performance is represented by a subscript “p.” Xp indicates the forecasted demand for units shipped/ received. Finally, requested performance during the period is represented by “r” subscript. This indicates the performance that the buyer actually required during the period, and may vary from what was planned or actually accomplished. Variance Definitions Total variance is the difference between the value of inventory actually available for use and the planned inventory value that was forecasted to be needed during the period. A positive total variance represents the value of inventory overages; a negative total variance represents the value of inventory shortages. Supplier performance variance is the inventory-related variance due to the inability of the supplier to perform as the buyer requested. A positive variance means actual value of inventory available for use when desired exceeded the request; a negative variance represents a shortage of availability. Supplier performance variance is the sum of product and time variances. Product supplier performance variance measures the difference between actual and requested inventory availability given the requested timing of that availability. A positive variance represents the value of inventory overages; a negative variance represents the value of inventory shortages. Product supplier performance variance is the sum of quantity and price variances.
Times supplier performance variance measures the difference between the value of shipments arriving on time and those not arriving as scheduled, given the actual value of inventory delivered. A positive variance represents the value of early shipments; a negative variance represents the value of late shipments. Quantity supplier performance variance measures the difference between actual and requested quantities of inventory available for use, given the requested price and timing of availability. A positive variance means too many units were delivered; a negative variance means that too few were delivered. Quantity supplier performance variance is the sum of condition and amount variance. Condition supplier performance variance shows the difference in inventory value between the percentage of product actually received in usable condition and the percentage requested in usable condition. A positive variance represents supplier performance that exceeded buyer request; a negative variance represents supplier performance that fell below buyer request. Assuming
TABLE 2 Supplier A Actual performance Shipment
Arrival
# of units
% Usable
Price/unit
1 2 3 4 5
on-time on-time late late late
500 500 600 560 400
100 80 90 100 100
$10 $10 $10 $10 $10
TABLE 1 Variable Definitions Variable X Y T P Q N
60
Summary data Definition Number of units shipped/received Percentage of units received in usable condition Percentage of shipments arriving on schedule Price per unit Quantity available for use (Q ⫽ X ⫻ Y) Value of inventory available for use (N ⫽ Q ⫻ P)
T X Y P Q N
Actual
Requested
Planned
40% 2,560 93.75% $10 2,400 $24,000
100% 2,500 100% $10 2,500 $25,000
50% 2,500 95% $10 2,375 $23,750
New methods of evaluating channel performance are needed. that a buyer might plan for, but not request, product delivered in unusable condition, positive variances should not occur. Amount supplier performance variance indicates the difference between the actual and requested quantities of inventory received, regardless of usability. A positive variance indicates that excess units were received; a negative variance indicates that too few units were received. Buyer planning variance is the inventory-related variance due to the inability of the buyer to forecast inventory availability needs during the period. A positive variance means requested value of inventory available for use when desired exceeded planned; a negative variance means a reduction from planned availability. Buyer planning variance is the sum of product and time variances. Product buyer planning variance measures the differ-
TABLE 3 Supplier B Actual performance Shipment 1 2 3 4 5 6 7 8 9 10
Arrival
# of units
% Usable
Price unit
on-time on-time on-time on-time on-time unscheduled unscheduled unscheduled unscheduled unscheduled
500 500 500 500 500 400 400 400 400 400
100 100 100 100 100 100 100 100 100 100
$15 $15 $15 $15 $15 $25 $25 $25 $25 $25
Summary Data
T X Y P Q N
Actual
Requested
Planned
50% 4,500 100% $20 4,500 $90,000
50% 4,500 100% $17.50 4,500 $78,750
75% 2,500 95% $15 2,375 $35,625
ence between requested and planned inventory availability, given the planned timing of that availability. A positive variance represents the value of unanticipated inventory demand; a negative variance represents the opportunity cost associated with investments in excess inventory resulting from over-forecasted demand. Had the planned forecast not exceeded actual requests, the dollars budgeted for inventory purchase could have been used elsewhere. Product buyer planning variance is the sum of quantity and price variances. Time buyer planning variance measures the difference between the value of shipments planned to arrive on time and those requested to arrive when scheduled, given requested availability. A positive variance means the buyer requested greater delivery accuracy than it had planned. Assuming that the buyer will request that all deliveries be made on time, a positive variance should be viewed as an opportunity cost. Budgeted investment in excess inventory availability could have been used elsewhere. A negative variance represents a reduction in the number of shipments expected to arrive on time. It may result from requests for expedited or additional unscheduled deliveries. Quantity buyer planning variance measures the difference between requested and actual quantities of inventory available for use, given the requested price and timing of availability. Positive and negative variances represent under- and over-forecasting of demand, respectively. Condition buyer planning variance shows the difference in inventory alue between the percentage of product requested in usable condition and the percentage planned in usable condition. As with the time buyer planning variance, assuming the buyer requests all product delivered in usable condition, a positive variance should be viewed as an opportunity cost. A negative variance should not occur since that would indicate the buyer had requested unusable product. Amount buyer planning variance indicates the difference between the requested and planned number of units of inventory received, regardless of usability. A positive 61
FIGURE 2.
Total System Variance–Supplier A.
variance indicates that the buyer requested a greater number of units than planned; a negative variance indicates a reduction from plan.
EXAMPLES The information in Tables 2 and 3 can be used to demonstrate the use of the framework. Suppose a buyer purchases product from two different suppliers. At the beginning of the period, the buyer’s plan is to purchase a total of 2,500 units of product A from supplier A to be delivered in 5 shipments of 500 units each. The planned price per unit is $10. The buyer requests that all inventory be received on time and in usable condition, but plans for 50% on-time delivery and 95% usability. supplier A delivers the first two shipments on time, but the last three are late. Supplier A also has problems delivering units in usable condition and, as a result, ends up delivering excess units. Table 2 shows actual performance and input variables for actual, requested, and planned performance. 62
The buyer also plans to purchase 2,500 units of product B from supplier B at a price of $15 per unit. Five shipments of 500 units each are planned. The buyer requests 100% usability and on-time delivery, but plans for 95% usability and 75% on-time delivery. During the period, it becomes apparent that the buyer’s forecast has significantly underestimated demand. As a result, the buyer requests 5 additional shipments of 400 units, each to be delivered on an unscheduled basis. Thus, the requested on-time delivery drops to 50%. The buyer requests a price of $20 per unit for these additional shipments, but because of the cost of expedited production and delivery, the actual cost is $25 per unit. Actual performance and input variables for actual, requested, and planned performance are shown in Table 3. The results of using the proposed framework to analyze the performance of suppliers A and B are shown in Figures 2 and 3, respectively. Looking just at total variance (planned vs. actual), it appears that supplier A outperformed supplier B. Yet, by examining more detailed levels of the analysis, it becomes apparent that not only did supplier B have superior
FIGURE 3.
Total System Variance–Supplier B.
performance but that the buyer’s inaccurate planning was a significant source of variance in the channel. The analysis for supplier A shows that the supplier did not perform as requested. The results for Supplier B clearly show that the buyer caused the majority of variance in the system. Looking at Figure 2 it can be seen that the buyer had planned excess inventory availability of $13,125. This should be viewed as an opportunity cost–the buyer has built a great deal of slack into its plans–since the budgeted excess could have been used elsewhere. IMPLICATIONS The use of this framework has several implications for supply chain management. First, the proposed heirarchical method provides increasing levels of detail about sources of variances that demonstrates the mitigating nature of variances throughout the supply chain. For example, the delivery of excess inventory (positive supplier amount variance) may mitigate the delivery of unusable inventory (negative supplier condition variance). Likewise, er-
rors in buyer planning may be offset by supplier willingness to exceed expected performance standards. Thus, management’s ability to isolate and correct the causes of variances is enhanced. In addition, it highlights which partners in the supply chain may be compensating for errors of others and should, then, provide information useful for strengthening the performance of the entire channel. Second, rather than using a more traditional and static means of performance analysis where actual performance is compared to that planned for at the beginning of the period; the model provides a means of performance analysis that is responsive to changing conditions in the marketplace. This is done by comparing actual supplier performance to buyer requested performance. Thus, how well a supplier met the buyer’s actual needs during the period rather than the needs expected at the beginning of the period is evaluated. In turn, this allows determination of how flexible suppliers are in responding to shifting buyer needs—and, by extension, the flexibility of the entire supply chain in responding to changing or unforeseen market conditions. 63
Finally, the framework provides a means of evaluating planning effectiveness. Because there will always be unforeseeable events in the market, it is unlikely that all planning variances could ever be driven to zero in a given supply chain. However, by highlighting where the planning process was inadequate, management cues for improving the planning process are provided. By not focusing exclusively on performance variances, the model contributes to the management of the supply chain by clearly showing the affect of planning on its efficient operation. REFERENCES 1. Lambert, D. M., Cooper, M. C., and Pagh, J. D.: Supply Chain Management: Implementation Issues and Research Opportunities. The International Journal of Logistics Management 9(2), 1–19 (1998). 2. LaLonde, B. J., Cooper, M. C., and Noordeweier, T. G.: Customer Service: A Management Perspective. Council of Logistics Management, Oak Brook, Illinois, 1998. 3. Ellram, L.: Total Cost Modeling in Purchasing. Center for Advanced Purchasing Studies, Phoenix, Arizona, 1994. 4. Ellram, L.: Total Cost of Ownership: Elements and Implementation. International Journal of Purchasing and Materials Management 29(4), 3–9 (1993). 5. Lambert, D. M., and Stock, J. R.: Strategic Logistics Management. Irwin, Homewood, Illinois, 1993.
64
6. Hewitt, F.: Supply Chain Redesign. International Journal of Logistics Management 2(5), 1–9 (1994). 7. Mabert, V. A., and Venkataramanan, M. A.: Special Research Focus on Supply Chain Linkages: Challenges for Design and Management in the 21st Century. Decision Sciences 29(3), 537–552 (1998). 8. Shank, J. K., and Churchill, N. C.: Variance Analysis: A ManagementOriented Approach. The Accounting Review 52(4), 950–957 (1977). 9. Hulbert, J. M., and Toy, N. E.: A Strategic Framework for Marketing Control. Journal of Marketing April, 12–20 (1977). 10. Shank, J.D., and Govindarajan, V.: Strategic Costs Analysis: The Evolution from Managerial to Strategic Accounting, Irwin, Homewood, Illinois, 1989. 11. Bentz, W. F., and Lusch, R. F.: Now You Can Control Your Product’s Market Performance. Management Accounting January, 17–25 (1980). 12. Lusch, R. F., and Bentz, W. F.: A Market-Accounting Framework for Controlling Product Profitability, in S. J. Shapiro and V. H. Kirpalani, Marketing Effectiveness: Insights from Accounting and Finance, Allyn and Bacon, Inc, Needham, MA, 1984, pp 481–508. 13. Ratnatunga, J., Pike, R., and Hooley, G. J.: The Application of Management Accounting Techniques to Marketing. Accounting and Business Research 18(72), 363–370 (1988). 14. Ijiri, Y., Kinard, J. C., and Putney, F. B.: An Integrated Evaluation System for Budget Forecasting and Operating Performance with a Classified Budgeting Bibliography. Journal of Accounting Research 6(1), 1–28 (1968). 15. Schiff, M., and Lewin, A. Y.: The Impact of People on Budgets. The Accounting Review 45(2), 259–268 (1970). 16. Weber, M. M.: A Framework for Analyzing Sources of Variance in the Supplier–Buyer Relationship: Determining the Contribution of Buyer Planning and Supplier Performance to Total Variance. Journal of Marketing Theory and Practice 4(2), 61–70 (1996).
Mary Margaret Weber The purpose of this work is to provide a framework that can be used to diagnose the causes of inventory-related variances in the buyer–seller relationship, and to analyze the costs to the channel associated with these variances. A multi-echelon approach is used to highlight variances caused by both supplier and buyer in increasing detail. The framework’s usefulness lies in its ability to separate variances caused by inadequate planning from those caused by poor performance. © 2000 Elsevier Science Inc. All rights reserved.
Address correspondence to Mary Margaret Weber, Emporia State University, School of Business, 1200 Commercial Street, Box 58, Emporia, KS 66801. Tel: 316-341-5315. E-mail: [email protected]
Industrial Marketing Management 29, 57–64 (2000) © 2000 Elsevier Science Inc. All rights reserved. 655 Avenue of the Americas, New York, NY 10010
INTRODUCTION The logistics practices of the past 15 years have made it clearly evident that in highly competitive domestic and international markets, the development of strong channel relationships is essential for the development and maintenance of competitive advantage. This trend toward strong channel relationships is thoroughly discussed in the logistics literature, for example [1–5]. Much of this push for stronger and long-term channel relationships focuses on the provision of superior customer service through better inventory management policies. Thus, new methods of evaluating channel performance that explicitly detail the impact of each channel member’s performance on the entire channel are needed [6, 7]. In particular, it has
0019-8501/00/$–see front matter PII S0019-8501(99)00112-1
Poor inventory planning affects other firms in a distribution channel. become increasingly important to eliminate variance from the channel relationship. In other words, increased competitive demands necessitate performance at high levels with decreasing acceptance of error. The purpose of this work is to provide a framework that can be used to diagnose the causes of inventoryrelated variances in the buyer-seller relationship, and to analyze the costs to the channel associated with these variances. A multi-echelon approach is used to highlight variances caused by both supplier and buyer in increasing detail. The framework’s usefulness lies in its ability to separate variances caused by inadequate planning from those caused by poor performance. Multi-echelon approaches were developed independently by Shank and Churchill [8] and Hulbert and Toy [9]. The Shank and Churchill model, later extended by Shank and Govindarajan [10], suggests that an analysis of variance in sales and marketing activities can benefit by adding increasing levels of complexity to enhance insight into the causes of variances. Each subsequent level of the analysis reduces the likelihood that large offsetting variances might be misinterpreted as a small, possibly insignificant, variance. A weakness of this work is that comparisons are made only between planned and actual performance, without an attempt to incorporate changes in the plan during the period being analyzed. Thus, this work allows no distinctions to be made between variances caused by inadequate performance and those caused by poor planning.
MARY MARGARET WEBER is Assistant Professor of Marketing at Emporia State University. She received her Ph.D. in Logistics from The Ohio State University and her MBA from Virginia Polytechnic Institute and State University. Her current research interests include the impact that information technology is having on the competitive structure of retailing and on the logistics practices throughout the marketing channel.
58
Hulbert and Toy’s work, later extended by Bentz and Lusch [11] and Lusch and Bentz [12], incorporates the idea of a revised plan into a multi-echelon approach. The use of a revised plan to evaluate forecasting accuracy is based on the broadly accepted concept of flexible budgeting, and it provides a means of recognizing the opportunity costs that arise from errors in forecasting [13–15]. Planning errors become apparent through comparison between the original and revised plans. While comparisons between actual and planned results can still be made to evaluate management performance, use of a revised plan eliminates the ability to hide any slack that was intentionally or unintentionally built into the original plan. The result is that a planning variance can be used to distinguish between the actual and forecasted operating environments. A performance variance distinguishes between actual performance and the performance expected, given the actual, as opposed to forecasted, environment. Weber [16] applied the multi-echelon approach to the evaluation of inventory-related variances in the supplier– buyer relationship. The model that was developed used actual, planned, and requested quantities of inventory to isolate causes of inventory surpluses or shortages. The model successfully distinguished between variances caused by inadequate supplier performance and those caused by errors in buyer planning. However, a major weakness of this framework is its use of placeholder values representing a combination of stockout quantity and time; it does not clearly elucidate costs associated with inventory-related variances. The framework proposed in this paper incorporates actual inventory valuations. Its use of both planning and performance variances helps to determine channel member responsibility for system variance. It also demonstrates how another channel member may compensate for planning or performance errors by one channel member and the cost of that compensation. The increasing levels of detail provide management cues for action in reducing overall system variance.
The variance model improves management’s ability to isolate and correct the causes of those variances. FRAMEWORK FOR VARIANCES IN TOTAL INVENTORY VALUATION Some of the same assumptions used by Weber [16] also apply to this framework. These include: 1)analysis of one product and one supplier–buyer relationship at a time; 2)the buyer–supplier relationship is such that buyer forecasts for the given time period are shared with the supplier; and 3)all variances are viewed as undesirable. The framework, itself, is shown in Figure 1. Positive
FIGURE 1.
variances represent inventory surpluses or opportunity costs associated with planning for excess inventory. Negative variances represent inventory shortages.
Variable Definitions Variables used in the framework are shown in Table 1. In addition, all variables are subscripted. Actual performance during the period is represented by a subscript “a.”
Fatal system variance.
59
The flexibility of suppliers in responding to shifting buyer needs can be determined. Thus, Xa indicates the actual number of units shipped/received. Planned performance is represented by a subscript “p.” Xp indicates the forecasted demand for units shipped/ received. Finally, requested performance during the period is represented by “r” subscript. This indicates the performance that the buyer actually required during the period, and may vary from what was planned or actually accomplished. Variance Definitions Total variance is the difference between the value of inventory actually available for use and the planned inventory value that was forecasted to be needed during the period. A positive total variance represents the value of inventory overages; a negative total variance represents the value of inventory shortages. Supplier performance variance is the inventory-related variance due to the inability of the supplier to perform as the buyer requested. A positive variance means actual value of inventory available for use when desired exceeded the request; a negative variance represents a shortage of availability. Supplier performance variance is the sum of product and time variances. Product supplier performance variance measures the difference between actual and requested inventory availability given the requested timing of that availability. A positive variance represents the value of inventory overages; a negative variance represents the value of inventory shortages. Product supplier performance variance is the sum of quantity and price variances.
Times supplier performance variance measures the difference between the value of shipments arriving on time and those not arriving as scheduled, given the actual value of inventory delivered. A positive variance represents the value of early shipments; a negative variance represents the value of late shipments. Quantity supplier performance variance measures the difference between actual and requested quantities of inventory available for use, given the requested price and timing of availability. A positive variance means too many units were delivered; a negative variance means that too few were delivered. Quantity supplier performance variance is the sum of condition and amount variance. Condition supplier performance variance shows the difference in inventory value between the percentage of product actually received in usable condition and the percentage requested in usable condition. A positive variance represents supplier performance that exceeded buyer request; a negative variance represents supplier performance that fell below buyer request. Assuming
TABLE 2 Supplier A Actual performance Shipment
Arrival
# of units
% Usable
Price/unit
1 2 3 4 5
on-time on-time late late late
500 500 600 560 400
100 80 90 100 100
$10 $10 $10 $10 $10
TABLE 1 Variable Definitions Variable X Y T P Q N
60
Summary data Definition Number of units shipped/received Percentage of units received in usable condition Percentage of shipments arriving on schedule Price per unit Quantity available for use (Q ⫽ X ⫻ Y) Value of inventory available for use (N ⫽ Q ⫻ P)
T X Y P Q N
Actual
Requested
Planned
40% 2,560 93.75% $10 2,400 $24,000
100% 2,500 100% $10 2,500 $25,000
50% 2,500 95% $10 2,375 $23,750
New methods of evaluating channel performance are needed. that a buyer might plan for, but not request, product delivered in unusable condition, positive variances should not occur. Amount supplier performance variance indicates the difference between the actual and requested quantities of inventory received, regardless of usability. A positive variance indicates that excess units were received; a negative variance indicates that too few units were received. Buyer planning variance is the inventory-related variance due to the inability of the buyer to forecast inventory availability needs during the period. A positive variance means requested value of inventory available for use when desired exceeded planned; a negative variance means a reduction from planned availability. Buyer planning variance is the sum of product and time variances. Product buyer planning variance measures the differ-
TABLE 3 Supplier B Actual performance Shipment 1 2 3 4 5 6 7 8 9 10
Arrival
# of units
% Usable
Price unit
on-time on-time on-time on-time on-time unscheduled unscheduled unscheduled unscheduled unscheduled
500 500 500 500 500 400 400 400 400 400
100 100 100 100 100 100 100 100 100 100
$15 $15 $15 $15 $15 $25 $25 $25 $25 $25
Summary Data
T X Y P Q N
Actual
Requested
Planned
50% 4,500 100% $20 4,500 $90,000
50% 4,500 100% $17.50 4,500 $78,750
75% 2,500 95% $15 2,375 $35,625
ence between requested and planned inventory availability, given the planned timing of that availability. A positive variance represents the value of unanticipated inventory demand; a negative variance represents the opportunity cost associated with investments in excess inventory resulting from over-forecasted demand. Had the planned forecast not exceeded actual requests, the dollars budgeted for inventory purchase could have been used elsewhere. Product buyer planning variance is the sum of quantity and price variances. Time buyer planning variance measures the difference between the value of shipments planned to arrive on time and those requested to arrive when scheduled, given requested availability. A positive variance means the buyer requested greater delivery accuracy than it had planned. Assuming that the buyer will request that all deliveries be made on time, a positive variance should be viewed as an opportunity cost. Budgeted investment in excess inventory availability could have been used elsewhere. A negative variance represents a reduction in the number of shipments expected to arrive on time. It may result from requests for expedited or additional unscheduled deliveries. Quantity buyer planning variance measures the difference between requested and actual quantities of inventory available for use, given the requested price and timing of availability. Positive and negative variances represent under- and over-forecasting of demand, respectively. Condition buyer planning variance shows the difference in inventory alue between the percentage of product requested in usable condition and the percentage planned in usable condition. As with the time buyer planning variance, assuming the buyer requests all product delivered in usable condition, a positive variance should be viewed as an opportunity cost. A negative variance should not occur since that would indicate the buyer had requested unusable product. Amount buyer planning variance indicates the difference between the requested and planned number of units of inventory received, regardless of usability. A positive 61
FIGURE 2.
Total System Variance–Supplier A.
variance indicates that the buyer requested a greater number of units than planned; a negative variance indicates a reduction from plan.
EXAMPLES The information in Tables 2 and 3 can be used to demonstrate the use of the framework. Suppose a buyer purchases product from two different suppliers. At the beginning of the period, the buyer’s plan is to purchase a total of 2,500 units of product A from supplier A to be delivered in 5 shipments of 500 units each. The planned price per unit is $10. The buyer requests that all inventory be received on time and in usable condition, but plans for 50% on-time delivery and 95% usability. supplier A delivers the first two shipments on time, but the last three are late. Supplier A also has problems delivering units in usable condition and, as a result, ends up delivering excess units. Table 2 shows actual performance and input variables for actual, requested, and planned performance. 62
The buyer also plans to purchase 2,500 units of product B from supplier B at a price of $15 per unit. Five shipments of 500 units each are planned. The buyer requests 100% usability and on-time delivery, but plans for 95% usability and 75% on-time delivery. During the period, it becomes apparent that the buyer’s forecast has significantly underestimated demand. As a result, the buyer requests 5 additional shipments of 400 units, each to be delivered on an unscheduled basis. Thus, the requested on-time delivery drops to 50%. The buyer requests a price of $20 per unit for these additional shipments, but because of the cost of expedited production and delivery, the actual cost is $25 per unit. Actual performance and input variables for actual, requested, and planned performance are shown in Table 3. The results of using the proposed framework to analyze the performance of suppliers A and B are shown in Figures 2 and 3, respectively. Looking just at total variance (planned vs. actual), it appears that supplier A outperformed supplier B. Yet, by examining more detailed levels of the analysis, it becomes apparent that not only did supplier B have superior
FIGURE 3.
Total System Variance–Supplier B.
performance but that the buyer’s inaccurate planning was a significant source of variance in the channel. The analysis for supplier A shows that the supplier did not perform as requested. The results for Supplier B clearly show that the buyer caused the majority of variance in the system. Looking at Figure 2 it can be seen that the buyer had planned excess inventory availability of $13,125. This should be viewed as an opportunity cost–the buyer has built a great deal of slack into its plans–since the budgeted excess could have been used elsewhere. IMPLICATIONS The use of this framework has several implications for supply chain management. First, the proposed heirarchical method provides increasing levels of detail about sources of variances that demonstrates the mitigating nature of variances throughout the supply chain. For example, the delivery of excess inventory (positive supplier amount variance) may mitigate the delivery of unusable inventory (negative supplier condition variance). Likewise, er-
rors in buyer planning may be offset by supplier willingness to exceed expected performance standards. Thus, management’s ability to isolate and correct the causes of variances is enhanced. In addition, it highlights which partners in the supply chain may be compensating for errors of others and should, then, provide information useful for strengthening the performance of the entire channel. Second, rather than using a more traditional and static means of performance analysis where actual performance is compared to that planned for at the beginning of the period; the model provides a means of performance analysis that is responsive to changing conditions in the marketplace. This is done by comparing actual supplier performance to buyer requested performance. Thus, how well a supplier met the buyer’s actual needs during the period rather than the needs expected at the beginning of the period is evaluated. In turn, this allows determination of how flexible suppliers are in responding to shifting buyer needs—and, by extension, the flexibility of the entire supply chain in responding to changing or unforeseen market conditions. 63
Finally, the framework provides a means of evaluating planning effectiveness. Because there will always be unforeseeable events in the market, it is unlikely that all planning variances could ever be driven to zero in a given supply chain. However, by highlighting where the planning process was inadequate, management cues for improving the planning process are provided. By not focusing exclusively on performance variances, the model contributes to the management of the supply chain by clearly showing the affect of planning on its efficient operation. REFERENCES 1. Lambert, D. M., Cooper, M. C., and Pagh, J. D.: Supply Chain Management: Implementation Issues and Research Opportunities. The International Journal of Logistics Management 9(2), 1–19 (1998). 2. LaLonde, B. J., Cooper, M. C., and Noordeweier, T. G.: Customer Service: A Management Perspective. Council of Logistics Management, Oak Brook, Illinois, 1998. 3. Ellram, L.: Total Cost Modeling in Purchasing. Center for Advanced Purchasing Studies, Phoenix, Arizona, 1994. 4. Ellram, L.: Total Cost of Ownership: Elements and Implementation. International Journal of Purchasing and Materials Management 29(4), 3–9 (1993). 5. Lambert, D. M., and Stock, J. R.: Strategic Logistics Management. Irwin, Homewood, Illinois, 1993.
64
6. Hewitt, F.: Supply Chain Redesign. International Journal of Logistics Management 2(5), 1–9 (1994). 7. Mabert, V. A., and Venkataramanan, M. A.: Special Research Focus on Supply Chain Linkages: Challenges for Design and Management in the 21st Century. Decision Sciences 29(3), 537–552 (1998). 8. Shank, J. K., and Churchill, N. C.: Variance Analysis: A ManagementOriented Approach. The Accounting Review 52(4), 950–957 (1977). 9. Hulbert, J. M., and Toy, N. E.: A Strategic Framework for Marketing Control. Journal of Marketing April, 12–20 (1977). 10. Shank, J.D., and Govindarajan, V.: Strategic Costs Analysis: The Evolution from Managerial to Strategic Accounting, Irwin, Homewood, Illinois, 1989. 11. Bentz, W. F., and Lusch, R. F.: Now You Can Control Your Product’s Market Performance. Management Accounting January, 17–25 (1980). 12. Lusch, R. F., and Bentz, W. F.: A Market-Accounting Framework for Controlling Product Profitability, in S. J. Shapiro and V. H. Kirpalani, Marketing Effectiveness: Insights from Accounting and Finance, Allyn and Bacon, Inc, Needham, MA, 1984, pp 481–508. 13. Ratnatunga, J., Pike, R., and Hooley, G. J.: The Application of Management Accounting Techniques to Marketing. Accounting and Business Research 18(72), 363–370 (1988). 14. Ijiri, Y., Kinard, J. C., and Putney, F. B.: An Integrated Evaluation System for Budget Forecasting and Operating Performance with a Classified Budgeting Bibliography. Journal of Accounting Research 6(1), 1–28 (1968). 15. Schiff, M., and Lewin, A. Y.: The Impact of People on Budgets. The Accounting Review 45(2), 259–268 (1970). 16. Weber, M. M.: A Framework for Analyzing Sources of Variance in the Supplier–Buyer Relationship: Determining the Contribution of Buyer Planning and Supplier Performance to Total Variance. Journal of Marketing Theory and Practice 4(2), 61–70 (1996).