Benefits of Applying Activity Based Costing in a Distribution Network
There has been considerable literature on Activity Based Costing (ABC) in the last 10 years. For those who are unfamiliar with this methodology a good place to start is to review the publications of Narcyz Roztocki of the University of Pittsburgh.
|
Activity Based costing in its simplest form is an accounting technique that allocates costs to products and services based on the activities of these products and services. It is one of several accounting methods that have been used over the years. Traditionally costs have been allocated based on an organizations structure, budget accounts or by some method of applying direct and indirect costs.
Many believe, including the author, that ABC when used correctly can provide a “truer” cost of the product or service provided to the end user. This is important as it will help to identify which products and services make/lose money and by how much. It also improves the strategic decision making process especially by identifying opportunities for both cost and price improvements.
Traditional costing systems tend to be based on historical information allocating overhead (logistics) usually on a volume-based measurement. ABC applies resource use directly to the output based on the actual work related activities that are required to produce the output with limited arbitrary allocations of indirect or overhead costs. The objective is to make indirect costs direct by tracing all expenses by an activity.
The objective here is not to debate the different accounting methods that are used to allocating costs but to show how ABC could be used in a distribution network. There are far more qualified individuals who have debated these issues in the trade journals. Each method has their advantages and limitations. It is sufficient for the purpose of this article that we agree that ABC when “properly applied” will provide a different picture of the actual cost of a product or service that may be more accurate than traditional accounting methods.
Current literature typically uses a manufacturing example to illustrate the possible results of using the different cost allocations. Traditional accounting allocates overhead normally by volume. ABC by measuring activities can focus on cross-functional processes that can help to identify non-value-adding activities and pin point the true drivers of cost at each stage in the process. The result is a more refined system of allocating overheads to products according to the demand by each activity. This focuses management’s attention on the underlying causes of cost.
This is especially important for logistics practitioners who want to show the direct impact of logistics costs to the bottom line. It emphasizes the importance of measuring the key performance indicators because “What gets measured gets done.” ABC improves the transparency of costs arising from many distribution channels and helps to identify the real profitability of market segments.
There are several misconceptions about ABC. First it is not a software package but a costing methodology that can be supported with a software package. This can improve the results by enforcing a proven discipline to the over all process. Second, many believe that ABC calculates “true” product costs and will result in better business performance. In reality ABC can provide a more accurate estimate of cost information but how it is used is still up to the manager.
For example, it is possible to impact sales and profits of an organization by identifying a more accurate price value relationship for products and services. Firms that use gross margin pricing many not be competitive in the market place because they are over charging for certain items. Conversely they may not be charging enough for other items which are utilizing scarce resources that could be reallocated to provide more profitable services that were not previously identified.
One of the issues with applying ABC is the level of effort required to implement a meaningful system. You must identify the key activities to be measured, the cost for these activities and than determine the cost drivers for each one. This can be time consuming and the data collection substantial. Too many activities and the results may not be worth the effort. Not enough and the accuracy may be suspect. Finding the proper balance is key to success.
Due to the complexity involved many organizations engage costly consultants and/or expensive software packages. These will certainly improve the result but in many cases are unnecessary. Some times it is better to be approximately right than precisely wrong. There are several ABC models available for measuring the costs of operating distribution networks. Those designed by accountants tend to be overly complex and have too many variables that do not add value to the process. Many logistics practitioners on the other hand are not sufficiently comfortable with the financial systems to be able to determine what are the key measurable cost drivers. A situation tailor made for hungry consultants.
Typical efficiency measurements of a distribution operation are cost per case, cost as a percentage of sales, cost per ton, cost per mile, etc. These are all valid and effective in managing the day to day operations and are necessary to maintain good cost control. What we also need for the strategic managing purchases are the costs to service specific customer, handle each product and service and ideally the cost by customer and product.
A simplified model that is currently in place tracks only three basic activities, order processing, selection of orders by customer and shipments of orders. This model was designed for a specific company but has been used elsewhere with some success. While it does not measure sufficient variables to ensure that all costs are allocated correctly it is directly accurate and will provide data that can improve the decision making process.
Let’s start with order processing. We all recognize that handling a customer order requires resources, time and equipment even if automated. The more orders the more resources needed to handle the volume. A simple method of identifying this cost is to divide the total cost of the order processing process department by the number of orders handled. The order-processing department can include customer service including the warehouse order coordinators and some informational technology. This has been estimated from anywhere from $50 to $125 per order. We will assume $50 for this exercise, which is probably low.
In many warehouse operations labor costs represent the largest single cost component and can be as high as 60% or more of the total budget. Typically, selection of orders is the largest component of the total labor cost. By measuring this activity it is possible to allocate costs to specific orders. The key insight is to identify the single most important activity in the operation and than measure it. This activity will vary by operation and as will be elaborated can be refined.
The model was first developed using a manual process by capturing a random sample of orders and measuring the actual time to select each one. The next step was to perform a regression analysis on the data testing the hypothesis that there is a direct relationship between order size and productivity rates. The cost allocation for each selection hour was estimated by dividing the total warehouse costs by the total selection hours, which we will assume, is $100 per hour. The result was the following productivity table:
|
Costs can now be allocated to each individual customer order by adding $50 for order processing and than an appropriate cost for order size using the table above. This can all be accomplished using spreadsheet software. In reality this process was further refined by sub diving the warehouse operation into clearly distinct departments. In this case it was dry grocery, dairy, produce, frozen and repack. Productivity tables were developed for each department using the same technique. Eventually the process was automated using engineered labor standards.
Allocating transportation costs can be relatively simple if you are using common carriers or it may be necessary to become more creative if you are using your own fleet. Common carriers typically charge by weight using a tariff by shipping lane. Accessory charges are also isolated by invoice and can be allocated by customer.
In this model it was necessary to allocate the costs for an in house fleet. This was accomplished by allocating costs based on drive time and unloads times to each customer. Similar to the warehouse, a cost for operating hour was calculated by dividing the total operating costs for the fleet by total drive and unloads times based on engineered standards. Where less than trucks load orders shared a vehicle costs was allocated by cube utilization as we normally cubed out before weighting out.
A similar model can easily be developed for allocating the logistics costs for an individual product. The key activities to measure are receiving, storage, selection, loading, transportation and an over head allocation. Productivity standards can be developed for receiving product based on cases per hour and the method of receiving such as “hand bombing,” pallets, etc. The hourly costs for the receiving crew, which normally includes the put-away function, can be estimated including the equipment. A cost per case is than calculated for each individual product. The process for loading trailers and allocating these costs is the same as the receiving exercise.
A simple method of allocating selection costs, at least initially, is to divide the total direct selection labor costs by the total cases selected resulting in a cost per case. A further refinement of dividing the warehouse into functional department similar to the first model described earlier is highly recommended.
An issue that needs further discussion is the location of the product in the warehouse and the potential affect that this could have on the cost allocation. Efficiency operators attempt to minimize travel time by locating fast moving items close to the receiving/shipping doors. This can reduce the handling time and ultimility the cost allocation for these items. As a general rule management decisions that can favour the cost allocation for an item over another should be avoided in the calculation. The costs should be socialized such as in the selection process described above. This avoids arguments above the placement and handling of products that could flavor one item over another that does not support maximizes the overall cost effectiveness of the total operation.
The allocation of storage costs normally creates some healthy debate. It involves two components, a cost based on the value of the product as well as physical storage. It also has the added complexity of wanting to allocate the costs by sales units. Costs such as insurance, over, shorts and damages and the cost on money are determined based on the value of the product. We are assuming that the cost of money is the actual costs of borrowing from a financial institutional or prime plus 1%. It could be argued that we should be using the opportunity cost of money or the companies hurdle rate as inventories are an investment against risk and should complete with other capital projects.
The following table will help to calculate the allocation of these costs.
|
||||||||||||||||||||||||||||
Assuming that the cost of carrying this inventory is 9.86% of the inventory value it is allocated over the sales units by the cost of the value of the individual unit divided by its inventory turns. (E.g. (9.86% * $5.00)/10= $0.049)
Similarly the physical storage costs need to be allocated over the total sales units. This includes rent, hydro, labor, physical labor, repairs, equipment etc. To simplify this exercise the value of physical storage can be calculated as what every is remaining after the previous allocations, excluding the cost of money which is not normally included in the warehouse budget. This would include over heads and provides a method of allocating this variable by a key activity. The cost per cubic foot is the result of the remaining physical storage value divided by the total actual cube of the inventory not the cubic capacity.
The following table shows some possible calculations.
|
||||||||||||||||||||||||||||||||
Assuming that the annualized cost per cubic foot of storage space is $4.81 than a case that is 0.10 cubic foot and turns 10 times will cost $0.048 per sales unit. Transportation is the only remaining activity and is allocated the same as in the first model. The final allocation by each unit is the sum of each of these activities. The third model cost by customer by product has yet to be developed but is believed to be feasible and may have already been completed by some enterprising logistics practitioner.
The objective of this article is not to provide a definitive model for applying ABC but to show how some companies have tacked the issue and to encourage debate on the subject. Supply chains in almost all industries are under increasing pressure to become more efficient and cost effective. Knowledge of the ‘true costs’ of a product or service (which includes all handling and distribution costs) is an essential tool to effectively embark upon supply chain management initiatives.
ABC is a useful tool that provides a means to measure and control key strategic priorities to reduce costs and develop economic incentives to encourage the end user to pull products though the distribution network improving the over all efficiency of the supply chain.
This article is a summary of a recent presentation made at the Canadian Association of Supply Chain & Logistics Management 35th Annual Conference in May of this year.