Procurement control for items with long transportation lead-time

Buffer Management principles work very well for procurement of short transportation lead-time items. However, additional controls are often needed for items where transportation lead-time forms a significant component of replenishment lead-time. A review of the buffer management principles would be instructive to understand this point.

Buffer Management principles

Let us look at the four Buffer Management principles under the Theory of Constraints.

1.  Initial Buffer between any two links in the supply chain = Maximum forecasted consumption within the reliable replenishment lead-time.

• Replenishment lead-time = Order lead-time + Production lead-time + Transportation lead-time
  • Order lead-time = Time between two consecutive orders or the time between the consumption of an item and placement of order.
  • Production lead-time = Time from the placement of order to the time that the supplier dispatches the order
  • Transportation lead-time = Time between the dispatch from supplier and the receipt of items in your warehouse.

2.  A buffer is divided into three equal zones demarcated by the colors red, yellow and green in that order starting from the bottom.

3.  Follow up and expedite the items when stocks are in the red zone.

4.  The buffer levels would have to be altered regularly because lead-time and consumption are prone to variability. Dynamic Buffer Management rule states that if the physical stocks of the purchased items in your warehouse are continuously in the red zone, there is always a threat that items would be stocked out and hence the buffer would need to be increased. However, if the stock is consistently in green, thereby indicating a high a stock, the buffer levels would have to be reduced. Enough evidence exists that an increase or reduction by a complete zone is enough to ensure availability of stocks without high inventories or excessive management attention.

The challenge

Shifting of supply base from the consumption location is increasing the transportation lead-times considerably. A typical case of procuring items by an Indian manufacturer from a Chinese supplier is illustrated below:

Order lead-time = 4 weeks

Production lead-time = 4 weeks

Transportation lead-time (On high seas) = 4 weeks

The resultant physical stocks in the warehouse in this case would be less than or equal to one-third of the buffer, i.e., the stock equivalent to the order lead-time. If the orders were to be expedited only after the stocks reached the red zone, it would be too late as the supplier would not have dispatched the items which would take another four weeks to reach the warehouse. The possibility of a delayed dispatch is high in this case since suppliers in China cater to a worldwide market and order size from Indian customers is comparatively very small. Hence resorting to action in the red zone is a sure-shot recipe for stock outs when the transportation lead-time is long.

The Solution

Since there cannot be any control on shipment time once the stocks are shipped, the point of control is prior to the shipment. There are two possible actions that can be taken to prevent stock outs:

1.  Requesting the supplier to ship much faster than the standard Production lead-time of 4 weeks.

2.  Use faster modes of transport such as air.

In other words, the control must be on the virtual buffer being the physical stocks and stocks in transit. The rules than are to be followed for preventing a stock out, therefore, are simple:

1.  Start requesting supplier to ship before the virtual buffer comes into the yellow zone.

2.  If the virtual buffer is in the red zone or nearing the red zone, chose a faster mode of freight.

Procurement managers are often assessed on material and transportation costs. Air freight is something they usually avoid. However in a majority of cases, air freight will be only a fraction of the money the company would make, had the items been available.

How to grow profits exponentially while improving sustainability?

Mahesh was in his office and putting together a document to convert the learning he has acquired to a powerful communication and implementation tool for his colleagues at work. He wrote the following on the white board:

• The goal of my organization is to always provide high quality products to its valued customers.  Needless to say that excellent communication and relationships with employees and suppliers is essential to achieve this goal. If these elements are taken care of, my organization will make profits now and in the future. My objective is to build an organization that consistently generates profits now and in the future.
• In terms of specific measurable results we need to make our profits three-fold in the next two years. Hence more products must flow faster from raw materials to finished goods and be sold to distributor.
• The number of customers who come to buy our products is limited. This is the most constraining point in the flow of products from raw materials to profits.
• Not having the products available when the customer demands it at a retail point leads to lost sales. The current mechanism of pushing products to distributors is causing our products to be unavailable at the retail point.
• Distributors are dissatisfied with the returns from this business. This is because they have high stocks of some SKUs that do not give them immediate cash and low, and sometimes zero, stocks of many SKUs leading to lost sales. They are unhappy with the balance between inventory and availability of SKUs.
• Distributor is our immediate customer. We can create a decisive competitive edge in the market if we can solve our distributor’s need much better than other suppliers.
• High availability of SKUs coupled with minimal inventory can be ensured for the distributor if we follow three simple rules.  First, we must supply according to the consumption by a distributor. Second, we must start with a stock level equal to maximum consumption within an average replenishment time multiplied by a factor for the unreliability of replenishment time.  This is called the buffer. Third, we must adjust the buffers by one-third if the stocks are too high or too low.
• We should convert transit warehouse at the plant to a plant warehouse. Introduction of a plant warehouse brings down the Finished Goods inventory while significantly improving the availability of products.
• The focus should not be on keeping every resource in manufacturing loaded 100% of the time. This policy leads to poor due date performance, long lead times and high WIP. We should instead focus on just responding to the demand signals from the plant warehouse.
• The buffers in the plant warehouse will be set up with replenishment time equal to half the current lead-time. Rule of adjusting the buffers will be the same as that defined for supplies between regional warehouse and distributor.
• The overall supply chain can expect many significant gains:
  • Distributor inventory turns will increase from 4 to 12 within three to six months.
  • The resultant sales increase from better availability, wider portfolio of SKUs at distributors and higher interest of the distributor will be at least 35% within a year.
  • Since our truly variable costs are only 50%, a 35% increase in sales can result in more than doubling our profits even assuming a 10% increase in non-truly variable costs.
  • In addition inventory in our supply chain will come down from current 52 days to about 30 days.

Mahesh was all prepared to communicate the above to his colleagues and was fully armed with the notes from the discussions he had been having with Greg over the past several weeks.

How do we reduce the Work-in-Progress in the plant while improving availability?

Greg was away to Mexico for a business engagement and was meeting Mahesh after over a month.

Hungry for intellectual stimulation, Mahesh resumed the discussions from where they left off the last time. “On the last occasion, we concluded that our policy that every resource must be loaded 100% at all times is leading to high Work-in-Progress (WIP) in the plant.”

“That’s right”, Greg said. “The flow of products is always choked at a few points in the chain of manufacturing. Let us call these points as constraints.  One of the significant constraints that you are facing is the flow of products from the distributors. You do not have any constraints in manufacturing or supplies. We, therefore, agreed that you should work towards satisfying the most significant need of distributors, i.e. inventory turns.”

Greg continued. “There are five solution elements when it comes to manufacturing. First, we need to decide at which point in the flow of products do we schedule orders. Secondly, what is the time required for the product to traverse from the raw material stage up to the point where the orders are scheduled in the flow of products. Thirdly, when should the material be released into the plant. Fourthly, how would priority be decided if two orders are waiting at the same point. Lastly, how do we dynamically manage the flow when the consumption and lead times are constantly changing?”

Mahesh was quick to respond, “As regards the first point, I think the scheduling of orders must be done at the points of constraint. The orders from the distributors will be serviced from the regional warehouse. The plant warehouse will in turn service the regional warehouse and the plant warehouse will trigger orders from the plant. As for your second point, I think the current manufacturing lead times are inflated because of urgent orders created by the absence of a plant warehouse and because of our policy of loading every resource 100% at all times. I guess, roughly about half the current lead times will be enough for the flow from the stage of raw material to that of finished goods.”

Greg added, “That is a great place to start. This would automatically bring down your WIP significantly. Hence the plant warehouse buffer level will be decided initially at maximum consumption within the replenishment lead time which is half the current lead time. You should not, therefore, release material till there is consumption from the buffer. Once the consumption or lead time changes, use the simple dynamic buffer management principle that we discussed earlier when we were debating distribution.”

Excited with the simplicity of the solution, Mahesh remarked, “I guess the answer to the fourth point is to give priority to the order which has penetrated much deeper into the buffer.”

Greg added, “Yes. As we discussed earlier, the buffer is divided into three equal zones color-coded as green, yellow and red. The operators in the plant can be given a simple rule that they should first focus on red orders before they move to yellow. Green orders should be taken once yellow is completed. If there are two orders of the same color let them use their own heuristic as to which order to be given priority. This is coined as Drum Buffer Rope (DBR) principle. Drum is the beat to which we operate, Buffer decides the length of time required and Rope decides when we should release the material.”

“Greg, I am going to implement this solution in our plant soon.  I might be picking your brains some time soon again!” said Mahesh.

“Sure buddy! Anytime!”

What is the underlying cause of high Work-In-Progress?

Greg decided to invite Mahesh to his office that Friday evening and continue their discussions.

Mahesh was impatient.

“Last week in your computation you assumed that a batch could be produced in 15 days. However we have about 45 days of Work in Progress (WIP). Hence I am assuming that on an average it takes 45 days for raw materials to get converted to finished goods. We have tried various methods to bring down the WIP but never found a lasting solution.”

Sensing the dilemma of a manufacturing head, Greg said, “Inventory is always an effect. To reduce the inventory, you need to solve an underlying conflict. What are the two important aspects to manage manufacturing well?”

Mahesh pondered for a while and replied, “During the beginning of the month, I am always asking the manufacturing head about efficiencies of each resource. At the end of the month I am always asking him to ignore efficiencies and meet the monthly plan. The first aspect is to manage efficiencies, i.e., eliminating all types of waste and leveraging every resource. The second aspect is to get products to flow through manufacturing.”

Extremely thrilled with Mahesh’s articulation of the issues, Greg said, “You have hit the nail on the head. What actions must be taken to achieve the differing needs of manufacturing system?”

“In order to manage waste, we need to look at efficiencies in each department and each machine. Similarly, to manage flow of products we need to look the global requirements of the company. Now I understand the dilemma. When we focus on global requirements, we ignore the need to manage waste. When we focus on managing local efficiencies, we ignore flow. But how do we ensure that both the needs are met?”, Mahesh asked.

Greg answered. “Now you have highlighted the dilemma of a manufacturer. Why do you think we need to manage local efficiencies to manage waste in the system?”

Mahesh was still analyzing, “My logic is each resource standing idle is a waste. Hence we need to get the maximum out of each resource.”

Greg nodded, “Great. Let us examine that assumption. Can you have all resources loaded 100% when there is a flow of products through a manufacturing system? What does it take to keep even one resource out of five in a manufacturing flow loaded 100% of the time.”

Mahesh quickly answered, “Assume that out of the five machines in the manufacturing flow, I want machine No.3 to be loaded fully. Machine No.2 can be down due to absenteeism and breakdowns. To compensate for the downtime of machine No. 2 and to ensure complete utilization of machine No. 3, machine No.2 should have excess capacities.”

Greg added, “Hence, to load machine No.3 fully, all preceding and subsequent machines must have excess capacities.  If we were focused on local efficiencies of every machine, then there would be lot of WIP in the plant.  If we need all resources 100%, then every machine needs lots of inventory waiting to be worked upon. This will increase the lead times, resulting in high Work in Progress (WIP) and poor customer delivery times.”

Mahesh concluded the discussion, “Now I understand that a resource standing idle is not waste. During the next week, can we decide on the processes needed for bringing down lead times in manufacturing?”

“Sure!” Greg said.

Why Plant Warehouse?

It was yet another session Mahesh was having with Greg.  This time the focus of the discussion was the need for a plant warehouse for Mahesh’s company.

Mahesh kicked off the discussion by raising an issue that was bothering him since their last meeting.   “During our last discussion you oversimplified the solution for providing superior inventory turns to distributors by assuming that all Stock Keeping Units (SKUs) are available in the plant warehouse. We do not have a plant warehouse currently. In fact, we have four regional warehouses that have about 45 days of stock. We, however, have a transit warehouse in the plant that can contain about 7 days of stock. I am worried that replenishing actual sales from regional warehouses will increase the overall inventories in the company. Besides, I am not sure if our plant is capable of supplying based on transfer from plant warehouse to the regional warehouse?”

Greg who was intently listening interrupted, “So does it happen that one regional warehouse has excess stocks of an SKU while another one has a shortage of the same.”

Mahesh responded that it happened all the time.

Greg thought for a second and asked, “Would you be happy with a solution that reduces the inventory in the company and at the same time supplies as per sales from your regional warehouses?”

“How is it possible?” asked Mahesh.

Greg started to explain, “Let me start with the situation you portrayed. There are SKUs that are in shortage in one regional warehouse and in excess in others. What would have happened if there was only a transit warehouse at the plant with no planned stocks?”

Mahesh responded, “I understand that there is always a difference between the forecast and the actual sales. If we had a plant warehouse that was supplying as per sales of regional warehouses, then there would neither be shortages nor excesses of inventory in the regional warehouses.”

Having got a satisfactory response, Greg moved on to address the next concern about inventory. “Now let us check if the inventory will be higher than the combined total of inventories in your four regional warehouses and the transit plant warehouse, i.e., inventory of 52 days. Could you now tell me how long would your plant take to produce one SKU?”

Having recently had a chat with the plant head on the same issue, Mahesh answered, “Even though the lead times may vary, it takes about 15 days on an average to convert an order to finished goods.”

Greg continued, “What makes up a large portion of the lead time is urgent orders that come from the regions where sales have overshot the initial estimate. This portion of the lead time would be eliminated with supply from the plant warehouse as per consumption. However, even if we take 15 days as the lead time, how much would be the inventory?”

Mahesh computed, “Since the lead time is about 15 days, we would need about 20 days of inventory to cover the variability of demand and the lead time.”

“What about regional warehouses?”, Greg asked.

Mahesh quickly computed, “There is a truck arriving at the regional warehouses every 2 days and the transportation time from the plant to these warehouses is about 4 days. Hence the replenishment time would be about a week and the inventory required to cover this replenishment time would be around 12 days. Are you saying that my Work in Process (WIP) and Finished Goods (FG) inventory would be in the range of 32 days instead of the 52 days we currently hold?”

Glancing at his watch Greg said, “Yes. You did the math. It is time for me to leave. When we meet next, we will chat about the solutions for manufacturing. Got to go now!”

What needs to be implemented to deliver superior inventory turns?

Mahesh and Greg met up once again at a snack bar near Greg’s work place on Friday evening.  It was a hectic, yet productive week for Mahesh and he thought the meeting to be a good way to end the week.

Having ordered samosas and tea, Greg picked up the thread from their last conversation.  He was intent on explaining how to ensure high availability of each Stock Keeping Unit (SKU) with minimal inventory.

Greg asked, “Let us assume that you have enough material in your plant warehouse. How would you supply your distributor?”

Mahesh responded, “I will supply distributors exactly what they have sold”.

Greg continued, “Would you incur additional transportation? Also, how would you determine the inventory that your distributors should hold at any given point of time?”

Sensing that the answers to Greg’s questions were lying in an unchartered territory, Mahesh replied, “I will monitor the daily consumption of each SKU for a distributor. When the total consumption equals my dispatch load, I will dispatch again.  Since the number of transportation loads remains the same, the transportation cost also remains the same. To respond to your second question, there must be enough stocks to last the replenishment time. Replenishment time includes transportation time, one-day taken to report the actual consumption and the time taken for accumulating a dispatch load. Since daily sales fluctuate, we may need to buffer it up by a certain percentage”.

Greg was impressed with Mahesh’s response. “The stock level you mentioned is called a buffer in management parlance. Buffer equals maximum consumption within average replenishment time multiplied by a factor for unreliability of replenishment time. Normally the buffer is divided into three equal zones and colour coded as green, yellow and red in that order from top to bottom.  My next question is, how would you adjust the buffer for either a change in consumption or replenishment time?”

Mahesh thought for a second and replied, “Maybe if the actual stocks are in red for one replenishment period, I would increase buffer levels. However, if the actual inventory is in green consistently, I would reduce the buffer levels”.

Greg added, “You can increase or reduce by one-third of the buffer. This has been proved to be adequate for taking care of even large fluctuations in consumption or replenishment time.”

Charged up by the discussions, Mahesh added, “Hence superior inventory turns can be delivered by observing a few rules. First, we must monitor daily consumption and supply when there is a dispatch load. Secondly, we create an initial buffer of maximum consumption with average replenishment time multiplied by a factor for unreliability of replenishment time. Thirdly, if the stocks with the distributor are in the red zone during one full replenishment time, we increase the buffer by one-third of the buffer. Similarly, we reduce the buffer by one-third if the stocks are continuously in the green zone.”

What is the most significant need of distributors?

Pursuant to the last soul-stirring conversation they had a week ago, Mahesh was eager to again meet with Greg.  Sipping his favorite Ethiopian coffee at Café Coffee Day, a much-mellowed Mahesh started from where they left the last time.

“Greg, I have been mulling over what you said during our last meeting that inventory turns is a very big lever to improve the Return on Investment (ROI) of distributors. I agree. But, is inventory turns a significant need of distributors?”

Greg was pleased by the fact that the ideas he voiced during their last meeting set his friend thinking. “You should know your distributors well. What do you think?”

Mahesh validated his point by adding, “I think inventory turns is a just an indicator of the problem that distributors face. Distributors’ business is investing money into products. They want faster rotation of money. But the real problem is that they are stuck with the wrong inventory.”

Greg said, “You are right. They have high stocks in terms of days of coverage for many Stock Keeping Units (SKUs). When there are high stocks, distributors face many negative repercussions.”

Mahesh continued, “I can think of several negative repercussions. First, the amount of cash the distributors have to invest in our SKUs is limited. High inventory of some SKUs leads to blockage of cash. This prevents the distributors from buying SKUs they need resulting in shortages and missed sales opportunities. Secondly, high stocks result in obsolescence when there is a product change. Distributors lose twenty times more than what they could have potentially earned when products become obsolete. Thirdly, they are constantly under pressure to offer discounts to liquidate stocks resulting in erosion of their margins. Fourthly, distributors reduce the range of products they buy. Hence they offer a lower range of products to the market thereby further reducing their potential sales. Lastly, they offer high days of credit to retailers.”

Greg pointed out, “While they have high stocks of some SKUs, they also have low stocks of other SKUs. There are several negative repercussions for low stocks as well. First, this leads to missed sales opportunities. Secondly, retailers tend to lose confidence in distributors and look for alternate sources. Thirdly, retailers will discredit the product in the market if there are repeated shortages leading to permanent impact on sales.”

Mahesh realized that resolving the said problem would have a significant bottom-line impact for his distributors.  “If we can resolve this, the distributors will have higher sales, lower inventory, larger product offering, less obsolescence, lower outstanding, better margins and higher retailer confidence. Wow!” Mahesh remarked.

Having contributed to the insight Mahesh just had, Greg summed up that when distributors had limited cash, having better availability with substantially lower inventory was their most significant need.

Before they parted, Mahesh ensured to make such meetings with his pal a frequent affair to benefit from similar nuggets of knowledge and extracted a promise from Greg to meet up again the following Friday.

Are we customer focused?

Mahesh and Greg were meeting after a gap of nine months. As usual, they needed some intellectual stimulation and the topic of discussion was, “Are we customer focused?”

Mahesh was the CEO of a 100 Million Dollar hardware manufacturing company. His company reached consumers through a distribution channel. He was clear that he would win this debate as this topic was very dear to his heart.

Greg, on the other hand, was a management consultant and had worked with many management leaders. He constantly challenged his own thinking and that of his clients.

Greg started the debate in almost a challenging tone, “Do you think you are customer focused?”

Mahesh was irked by the tone of Greg’s question, yet he responded nonchalantly, “Is there a doubt? We produce excellent products that meet the needs of our consumers. Our products are long-lasting. We also back it up with good after sales service. The fact that we have 35% market share in our products is a clear demonstration  that we are focused on our customers’ needs.”

Greg pondered for a second on this response and continued further, “Mahesh, who is your customer? Customer is someone who pays you for your product. In other words, your immediate customer is your distributor who pays you although he does not consume your products.”

Mahesh, who usually won most of his arguments, was getting edgy with the debate and replied grudgingly, “Agreed. But we do take care of  our distributors by giving them margins that are comparable to other players in the market.”

“I agree that your margins are comparable to that offered by your competitors. But is that what your distributors look for? What is the distributors’ need?”, Greg persisted.

Mahesh was clear that distributors were businessmen who wanted a return for their investment and that his company gave them good margins. Greg wanted to know if there were any factors besides profit margins that could  impact Return on Investment (ROI).

Wondering where the discussion was leading to, Mahesh answered, “Yes. ROI of distributors will be margins multiplied by the inventory turns.”

Greg asked, “What is the average inventory turns for your distributors?”

Having never viewed the distributors’ business from that angle, Mahesh started to look in the direction Greg was taking the discussions to and said,  “Maybe four turns. I think my distributors carry about 3 months of stock. With 5% margin retention, they make 20% returns on their cash invested in the business. In addition they have expenses which reduce their returns further. I agree that perhaps our focus is not on our distributors. How do we impact the inventory turns of our distributors?”

Greg, having gotten Mahesh’s attention, continued, “What you stated is just the result. We should analyze your practices that caused such a poor ROI for your distributors. There are a series of questions that I have for you. What do sales people get incentives for? How are their performance measured? What do they do during the month end when they have to achieve their targets?”

Mahesh saw the point Greg was trying to drive home and answered,  “Sales people are measured on the sales to the distributors. I have been in a sales role before and I can confidently say that the focus is usually to push stocks to distributors during the month end. We do that irrespective of the amount of stocks the distributors carry. Before you score a point, let me admit that sales people are not distributor focused.”

“Thanks for agreeing. Now there are few questions from a supply chain perspective. How does your manufacturing division plan its production? Do the distributors always get products they want?”. Greg was getting confident about his hypothesis.

Mahesh did not think that there was anything wrong with their planning, “Distributors often do not get Stock Keeping Units (SKU) they want. But the problem is with the distributors’ forecast. We produce what they forecasted. So we are not to blame if distributors’ forecast is bad.”

Greg knew this argument well. “Would you agree that if you had a better way to forecast demand, then you would have done a better job of customer satisfaction? I am sure you would.  Is forecast accuracy better at a distributor point or at an aggregate point like your plant?”

Mahesh looked happy with this new insight.  He said, “I agree that the accuracy is better at an aggregate point. Are you saying that our systems are not geared for customer satisfaction?”

Greg knew that he achieved his objective to get Mahesh thinking and said, “Mahesh, do consider the point I was trying to get at.  I have done work for another organization like yours. We improved the inventory turns of distributors in that company from 6 to 20 within 6 months. Inventory turns, in my experience, is a bigger lever to improve the ROI of distributors than margins. I got to leave now, but when we meet the next time, I will explain the steps needed for significant inventory turns.”