III
ENHANCING DISTRIBUTION



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 51
--> III ENHANCING DISTRIBUTION

OCR for page 51
This page in the original is blank.

OCR for page 51
--> American Red Cross Blood Distribution System Jeany Mark I would like to talk to you today about the American Red Cross as a total system. We have recently created a "Hub" to increase the availability of blood products by coordinating supply and demand through a consolidated inventory management. One of the things I have been doing this past year is to take a good system, and I believe it is a good system, remove some of the inefficiencies, and then create greater efficiency in what remains. Before I describe that, however, I would like to go over some of our very basic trends to give you a sense of our supply and demand situation with the Red Cross. We have looked at the average monthly distribution of blood and blood products from 1986 through 1995. The whole blood collections include everything: autologous, allogeneic, and directed. We have taken an average for the past five years and have found that our collections and distributions have matched. The real question here is whether demand drives our collection behavior or whether if we had collected more units of blood, we could have sold them. The other point that is of interest to our organization is the issue of platelet distribution. Let me just very briefly explain this. From our perspective, we are seeing a substitution effect. That is, as single-donor platelets use increases, random donor platelets usage declines. Interestingly, we used to be growing quite well in apheresis platelets, but it has really been leveling off for our system. If you look at the trend, the total platelet utilization is now tipping downward a little bit. Moving back to improving the availability of blood, we believe that in our system it is driven mainly by the market demand, blood typing needs, and the dating issue. The Red Cross whole blood collection has been going down. At one point we used to collect 6.1 million units. Today we are collecting 5.7 million units. However, I believe a lot of the downward trend is driven by autologous and directed donations. Allogeneic collections have been decreasing at a rate of 1/10 of 1 percent on a compound annual growth rate basis, whereas autologous and directed collections have been going down by about 1 percent.

OCR for page 51
--> As far as what can we do to increase the availability from a systems perspective, we are looking at coordinating the supply and demand through more of an inventory consolidation, that is, using the Hub concept. The purpose of the St. Louis Hub is to ensure the balance of supply and demand within the ARC as a whole, and to be a single coordinating point for transactions with non-Red Cross blood centers and hospitals. We have restock programs whereby importing ARC centers automatically receive blood products when their inventory levels reach a predetermined threshold. This also obviates the old practice of having to receive unneeded A-positive units in order to get needed O-positive units. Now people can call the Hub and get O's and B's without having to be penalized by having to include and pay for units of other blood that they don't need. The Hub also plays a role in lessening the impact of unanticipated sharp fluctuations, using some of the odd lots of smaller-unit-volume shipments that you would normally not be able to ship to a big importer such as Los Angeles without incurring a lot of transportation costs. We aggregate them all in the Hub. We have this information in a computer system, and the regions report their inventory level every day. The other issue that we have is trying to get all of our organizations to adopt uniform definitions of working inventory and critical inventory levels, so that supply can be centrally managed in a much better way than has been done in the past. Today there is still inconsistency in the definitions of these parameters. We now believe the optimal level is three days' worth of inventory. The emergency level is one day's worth. We have been hovering in between the two, that is, between optimal and emergency. There are days that it gets pretty close. I know how it feels when you have 10 units of O-positive on your shelf and you are taking care of 70 hospitals. I know a lot of our blood centers do start to panic then and they do call. But generally when it comes time to really help out, our regions do cooperate and help other regions in dire emergencies. I mentioned dating as one of the other areas that can increase blood availability. First of all, the restock program provides a much better shelf life. Instead of giving people 15 days to work with, we try to give them a minimum of 36 days. Part of the reason also is that we don't want to have to keep worrying about moving the inventory back and forth, having them order multiple times each week. They also function as a clearinghouse for short-dated products, that is, those units whose shelf life is near expiration. For instance, we get calls from some of our locations in Miami and San Juan requesting blood. We tell them we have some short-dated blood and they take it. We also know certain blood types are less in demand. The one thing that the Hub is trying to do is that instead of moving unneeded short-dated type A units all over the place, we will ship them all in and let them outdate at one

OCR for page 51
--> place. From a Red Cross perspective, whether it is outdating in Philadelphia or outdating in St. Louis, it doesn't matter to the organization. It is the same dollar amount. Overall, however, the Red Cross system outdates rate has gone down. I don't think that it is due to any particular reason. There are probably a number of reasons over time. As I mentioned earlier, we currently move blood all over the place, moving most of the red cells from exporting regions to the Hub and then redistributing them to the importing regions. It makes better sense, though, to do direct shipments as much as we can, basically from the exporters to importers. It just doesn't make sense for Philadelphia's imports to come from as far away as Boise, Idaho, when there is a Johnstown Blood Center in Pennsylvania that has the same volume available. We are therefore changing to a "virtual inventory" management system in which the bulk of the products bypass the Hub, which only coordinates and directs the movements. The other drawback of moving everything through the Hub is that we create an extra day of float that we don't need to do. The newer direct routing takes care of that, and we ultimately save some money, too, by eliminating double shipping. Our regions have now been regrouped into five areas: Western, South Central, Southeast, North Atlantic, and North Central. The idea is to get each of the areas to be as self-sufficient as possible because they do have a mix of importers and exporters, and where they are short, then we will go through the Hub. Although the Hub will be carrying far less physical inventory, we still want to carry some for emergency purposes, for instance, like the strike that we recently had in one of our locations. Another area I want to touch upon a little bit is the pricing issue. A year ago, if you looked at the intra-Red Cross movements, prices were all over the place. I would like some of our non-Red Cross friends out there to know that we sell blood cheaper to you than we do to our own partners within the Red Cross system. What we are doing now is standardizing the price. A unit of red blood cells will now be worth the same amount no matter what Red Cross center you buy it from or sell it to. There is no such thing as a markup, so that people won't waste time trying to make the best deal possible. We are focusing on matching supply rather than what is the best deal. As a result, we are in the process of eliminating all the internal billing, where it makes sense, and just do adjustments. Finally, there are many other future challenges for the Red Cross in its inventory management. One is that we need to have a much better system to manage and forecast the supply and demand. A second challenge is to better educate the local community about blood not necessarily being a strictly local resource. They may not be willing to help out other blood centers if we cannot overcome the notion that what's collected in their community is strictly for use in their community.

OCR for page 51
--> The third challenge is something that the Red Cross will be spending the next 18 months very aggressively working on is driving down some of our costs. I know that there is this issue of whether blood should be treated as a commodity, but I talk to hospital administrators and they tell me they don't care what we have to say, dollars and cents do matter. Questions/Comments Paul Russell: Ms. Mark, you are saying that there needs to be some kind of central coordination within the Red Cross to permit efficient transfer of blood from one distant place to another, and that is not necessarily through your Hub. Jeany Mark: That is correct. Right now we have a system where everything is actually shipped into St. Louis. We no longer really need to do that. When we have a standardized computer system, it will have a requirement that you can look up each region's inventory level. So, the concept is not to move it physically, but to go point to point. Paul Russell: Logically speaking, if your approach is as you have described it, then you should also have some kind of oversight about non-Red Cross sources. Are you planning that? Jeany Mark: Yes. We are talking with the few non-Red Cross partners we have and looking at supplying them. When we need to, we buy from them also. Paul Russell: Your concept ideally then is of a national system centrally coordinated to include everybody. Jeany Mark: Yes, that is correct. A week ago we sent out an internal communication to our Red Cross units directing them to refer all blood product transactions with non-ARC centers to the Hub. Alvin Drake: If hypothetically, then, Dr. Gilcher can supply Los Angeles far less expensively than any Red Cross center, may that happen today or does it have to go through the Hub? Jeany Mark: Through the Hub. We are trying to manage our own resources so that when there is a demand from outside the Red Cross system, then the Hub makes the call. For example, the Hub calls United Blood Services,

OCR for page 51
--> Oklahoma Blood Institute, or other non-ARC blood centers. Likewise, if the non-ARC centers have excess that they want to declare to the Red Cross system, the Hub is the central point for these non-ARC centers to contact. This is the thinking for now, but I see other possible alternatives that may be more effective and strategic to the Red Cross in meeting our customers' demands.

OCR for page 51
This page in the original is blank.

OCR for page 51
--> Blood Resource Sharing Programs Toby Simon I was very intrigued by the hub concept when the Red Cross put it together. It is interesting that they are now beginning to partially move away from it and toward direct transportation; at United Blood Services (UBS) we couldn't see the economic sense in the double transportation in this age of computers. We do have a central computer system for United Blood Services, so that we control inventory from a single coordinating position in Scottsdale, Arizona, but we don't physically move the blood into Scottsdale. We move it from its area of availability to its area of need. Like the Red Cross, we have two systems of moving the blood. One is by prearranged agreement. In each budget cycle, those centers that anticipate need provide an estimate of what they will need to import. Those centers that anticipate surplus indicate that, and we try to match it up and have regular commitments through the year. In addition, we also have on a daily basis the movement of blood from area of oversupply to area of need. That is done by a computer system on which the inventories are pulled up each morning, and the central inventory control person makes certain changes in the morning based the computer display of who has surpluses and who has deficits. Then as the day goes on, he coordinates spot needs that occur. It is on a spot basis that we provide blood outside the system, since we have a surplus and centers, Red Cross or non-Red Cross, elsewhere in the country have needs. We no longer have any special commitments. We believe that it is ethically and morally appropriate for blood to be shared as a national resource and to go from areas that can draw it more readily to those that have greater needs, whether it is because of a greater concentration of tertiary care centers or other factors. We do believe, however, that a center should never short its own community. So, we have made a special point in the last several years of assuring that any center independently can say, "No, we cannot ship. We cannot meet that commitment because of shortages in our own community." However, there is a strong feeling of camaraderie among our centers, so, just as was mentioned with the Red Cross,

OCR for page 51
--> people will make that effort, and so generally we meet those needs. We understand that there is something similar among the Blood Centers of America, a group of independent blood centers that are coordinated in a somewhat looser fashion out of Rhode Island. They have a single system and make both prearranged contractual arrangements through the year and spot arrangements to share blood among participating blood centers. The major national program is that of the American Association of Blood Banks (AABB), which runs a national blood exchange through its headquarters in Bethesda, Maryland. We participate in that program, which is actually highly efficient and works quite well. We utilize it very extensively, both when we have needs that somehow we can't fill and also when we have surpluses that we want to share with the rest of the country. There is a little controversy with this system, however, because the AABB will ship a unit, for example, from United Blood Services to a hospital in Boston that is presumably supplied by the American Red Cross or a hospital in New York that is presumably supplied by New York Blood Center. That is somewhat controversial for antitrust reasons and for reasons of the diversity of the AABB membership. The AABB believes that it is obligated to do so on the basis of the fact that these hospitals are members and have qualified for the National Blood Exchange. The antitrust implications stem from the possibility of anticompetitiveness charges if member hospitals are prevented from utilizing this system in lieu of their local blood center. We can talk about that a little more in detail later, but the National Blood Exchange has received a major commitment from the AABB to make blood available as a national resource throughout the country, and many people believe it is highly efficient and highly effective in moving blood from oversupply to undersupply. The Council of Community Blood Centers (CCCB), of which I am currently president, also runs a program, but it is basically a simple faxnet type of program, whereby if somebody needs blood, it goes over the faxnet and another CCBC member can reply and make the provisions. In both the AABB and the CCBC, the cost of blood from one center to the other is largely determined by the costs that prevail in the supplying center, and there is a small transactional cost with the National Blood Exchange in order to keep the system afloat. The National Blood Exchange is a managed system with professional management, and the CCBC system right now is not. American blood policy, which was promulgated in the early 1970s, had two major points, which we have continued to discuss even today at this meeting: regionalization, the concept that there should be a bringing together of hospital needs and supply on a regional basis, and voluntarism. The rational sharing of the national blood resource in a way is an extension of the regionalized concept. If we do have a national blood resource, where does the blood come from?

OCR for page 51
--> I think what we are beginning to see is a coming together of all of the things that we are discussing today because, obviously, you cannot have a sharing if you don't have a collection and an adequate recruitment program. We know that it comes to some extent from rural areas, and I presume that there may be several reasons for this. One is that rural areas have fewer tertiary care centers, so they do not have the transplants or the trauma and may not be offering neonatal intensive care and other kinds of medical care that require large amounts of blood. They also may have a greater sense of community. We used to have our most successful blood drive in New Mexico in the community of Los Alamos, which is where the atomic bomb was developed. It is 60 miles north of Santa Fe, on a mountain. Once I was telling one of our friends who had lived there what a wonderful thing it is that the community does. "Well," she said, "of course, it is the most exciting thing that happens there all year." Our most successful recruitment programs are in the Dakotas and Minnesota, very much like the Red Cross. If you look at our Fargo, North Dakota, program, it is highly effective, highly cost-efficient, and the source of a predictable blood supply throughout the year. They use very little fixed-site draw and therefore have lower levels of donor retention than one would expect in this area of the country. Their draw is very heavily based on bloodmobiles in rural areas, which you would think would be costly because you are sending teams out and they stay in hotels. Because that blood supply is so reliable, they actually have an extremely low cost of collection and a very effective program. On the other hand, our Phoenix blood center has been setting records emphasizing donor retention and fixed sites and reducing their mobile blood drives. Different things work in different areas. We know that rural areas in North Dakota do better than rural areas in Mississippi, Louisiana, and Texas. We assume that some of this has to do with higher educational standards, lower unemployment, and so forth. Ethnic homogeneity and acculturation are issues that also enter into the equation, and that can be a little bit difficult to discuss, but many of the areas that have been pinpointed as areas of ready supply tend to be more ethnically homogeneous. Perhaps this is related to people's feelings about contributions to the community or sense of community. One of the interesting points that we can make from United Blood Services concerns our program in New Mexico. The state has the highest ethnic minority population in the country—50 percent Hispanic American and Native American population. While not the highest contributor in the United Blood Services, the program in New Mexico does produce a surplus and shares with other areas of the country. When I discussed the New Mexico donors with our recruiters in the Rio Grande Valley, I told them that obviously one can be successful with Hispanic donors. The recruiters, however, point to the

OCR for page 51
This page in the original is blank.

OCR for page 51
--> Supply and Demand in Transfusion Services Arthur Bracey I come today as a user of blood. In fact, at my facility we transfuse about 50,000 components a year. What I will try to do is to give you the transfusion service perspective on the issues of supply and demand. We should know, based upon the blood orders that come into the transfusion service, what our inventory needs are. Unfortunately, we aren't in a system where we get blood orders far in advance. In elective surgery, we find out about the surgery schedule sometimes 6 hours, sometimes 12 hours ahead of time. If we could find out further in advance when elective surgery is scheduled, we could have a better-coordinated system for projecting daily blood demands. In addition, we have inventories of blood in our hospitals, which are known largely only to the hospitals and are often unknown to the blood centers. In this day of computerization, one would think that in many centers the hospital inventories could be monitored on a periodic basis with an electronic process that allows data to be fed into the regional blood center for assessment of regional inventory. In fact, such a system could facilitate estimation of the national inventories and resource sharing. The problem is that the exchange of information is not routine. In my area exchange of inventory data is crisis driven. I find out about a blood shortage when it is there. I don't find out about an impending blood shortage. Part of the problem is also that we are stuck with old techniques. We still exchange data primarily by telephone conversation, not by computers. In a transfusion service, we often have marked fluctuations in blood needs. For instance, we had a case on Wednesday of a woman who had a placenta previa and suddenly dropped from a hemoglobin of 14 g/dl to a hemoglobin of 5 g/dl. The prospect of such an urgent situation demands that we stockpile blood in order to be sure of meeting all patient needs. The question that comes up is how efficient are we at determining the stockpile? We have talked today about meeting the inventory needs of transfusion services, but no one has talked about critically assessing how that need is determined. Important factors in determining inventory requirements include the institution's approach to providing blood for patients, that is, the

OCR for page 51
--> crossmatch-to-transfusion (CT) ratio, the length of time units of blood are held in reserve, and other factors that are management issues from the perspective of a hospital transfusion service. Unfortunately, by virtue of its perishable nature, blood has a limited life span that affects its availability. There has been work done largely out of the New York Blood Center addressing inventory modeling based upon evaluation of wastage versus shortage of blood products. If we look at my institution's usage on a day-to-day basis (see Table 4 ), there are peaks and valleys. On the weekends, we need much less blood than we need on routine workdays. Our surgeons like to operate Monday through Wednesday. They take off early on Friday. Saturday and Sunday are low-volume days. Thus, we have data that allow us to anticipate when our peaks and valleys will occur. We look at the efficiency of the use of inventory as a function of outdate, but is that really the correct answer? Is that the correct approach to determining the efficiency of inventory management? TABLE 4: Red Blood Cell Use (units) at St. Luke's Episcopal Hospital by Day of the Week Week Number Mon Tues Wed Thurs Fri Sat Sun 1 101 68 45 55 37 20 42 2 69 68 123 66 37 34 67 3 75 78 69 63 46 24 49 4 88 105 103 54 51 17 27 5 61 65 68 69 42 30 70 6 98 115 88 107 41 32 62 7 59 75 69 74 38 29 52 8 105 60 76 116 47 19 120 Mean 82 79 80 76 42 26 61 Std Dev 17 19 23 22 5 6 26 Inventory is assessed largely by blood wastage and blood shortage data, that is, inability to support a patient need. Ideally, the inventory requirements should be set by historical transfusion volume and case demand. I am not sure, however, how often that happens. A real challenge that we face in transfusion medicine is that of improving our efficiency of inventory management. I know that a person having a cardiac bypass procedure will use 2.2 units. I am trying to learn how much blood a nonsurgical patient with

OCR for page 51
--> lung cancer will use. We have been most successful at predicting the surgical blood need, but I think we have a lot to learn as far as estimating nonsurgical blood need and nonsurgical case volume. I want to emphasize that many studies assessing surgical blood support have been published, but many facilities are not using them in determining inventory levels. Sinelson and Bradheim's work on inventory management is interesting.6 Their model is based upon analysis of the normalized stock level, which is a function of blood stock divided by the mean daily usage. They developed a formula to predict the interaction of outdate and shortage based upon selected normalized stock levels. The formula, in fact, when applied to real hospitals, proved accurate in predicting shortage and outdate. The problem here is that you have to have set points for acceptable frequencies of outdate and shortage. In routine transfusion service practice, many directors have no input regarding these set points and may be unaware of any such determinations by others. From the perspective of a transfusion service, the blood centers which could be helpful are often underinvolved in helping to set transfusion service blood inventory levels. If the director of a transfusion service wants to be safe and avoid any blood shortages, he can just stockpile enough blood to be sure he will never run out. Of course, as you decrease the amount of the normalized stock ratio, which is the amount of stock related to the mean daily demand, there will be an increasing frequency of shortage. There is another important consideration. Blood that is out in the field may not be being used appropriately. We feel we have shortages on the regional and national levels, but we often have maldistribution of blood rather than true shortage. The places that have a small mean daily transfusion have a real problem in avoiding excessive blood outdating. Likewise, it is inefficient for a hospital to keep blood crossmatched for three days. Thus, in hospitals it is very important to look at issues of crossmatch-to-transfusion ratio. In the transfusion service, we can't really impact the shelf-life of the blood, but we certainly can manage certain variables, such as how long we will keep blood crossmatched. When you have centralized facilities where transfusion support care of many patients is taking place, you have economies of scale which mitigate against blood maldistribution. With respect to shortage, there is a convergence in the curves predicting shortage frequency as you keep less blood. If you have a normalized stock ratio of about 1.5, you begin to experience a significant amount of shortage, but, again, what is shortage? Shortage is having to call the blood center to ask it to provide blood. If more blood is available at the blood center and the 6   Sinelson, V and E Bradheim (1991). A computer planning model for blood platelet production and distribution. Computer Methods and Programs in Biomedicine, 35: 279–290.

OCR for page 51
--> transportation system is efficient, then this situation really isn't a problem since it doesn't impact patient care. However, this scenario demands an efficient transportation system. If I need a unit of blood urgently, I don't want to hear that it is going to come in two hours. I would rather hear it will be delivered in 15 minutes. From the perspective of the transfusion service, I don't want to call various hospitals or regional blood centers when I need support. I want one number to call and I want the blood in 20 minutes. That is all I want. Prior to the era of Good Manufacturing Practices (GMP) as directed by the FDA, blood was available at an earlier time following collection. In my area, by 8:00 A.M. we could inventory blood that had been drawn on the preceding day. Post-GMP, the blood is available at noon, but what surgery starts at noon? This delay is particularly an issue with units that have a shorter shelf-life, such as platelets. Will centralized donor testing further decrease the responsiveness of the blood centers to provide that urgent need that we have? If you don't have a good transportation system, there will be unnecessary stockpiling. So, the system needs to be focused on rapid and efficient provision of blood products. What about remote depots? That was tried in our region, and it cost more because someone had to man the post 24 hours a day. I don't know if it could work, except maybe in places such as Alaska, where remote depots may be the only acceptable alternative. When there is a shortage in a blood transfusion service, you are always wondering in the back of your mind if the first order always gets delivered first. I still don't know that answer. Another question is whether blood orders are related to real need. Are the inventories fairly and equitably distributed? Again, what is most important is speedy and reliable delivery. Blood products have got to be there when you need them. Return policy is another issue. I understand a blood center discouraging return of short-dated blood, but I just don't understand some return policies. Currently in our region, if you order leukocyte-reduced blood, you can't return it, despite its long shelf life. I don't understand the reason for that policy when you can return other sorts of blood. Some policies impede the free interchange of blood, which is important in facilitating distribution. There is also the issue of the delivery fee. If you are going to charge large sums to have blood delivered, then people will stockpile the blood, potentially keeping more blood available than they need. Blood products need to be made readily available and transported without exorbitant fees. There was a study in Sydney, Australia, of a blood center's role and what it could do when it got involved with the hospitals in the region to improve inventory management. The blood outdate rate dropped from 5 percent to 0.9 percent after implementation of an educational effort. Factors that were found

OCR for page 51
--> to be important in this particular study included: (1) whether or not the hospital had a functioning rather than a perfunctory transfusion committee; (2) smaller hospitals had greater problems with outdates and management of inventories; (3) CT ratio greater than 2 was found to be a significant problem with respect to outdate; and (4) those institutions that were remote kept more blood and outdated more blood. The Wall Street Journal recently reported that 20 percent of hospitals that transfuse autologous blood have mistakenly given allogeneic blood prior to transfusing the intended autologous blood. These errors occur because we often have these special requests, but no fail-safe way to link the patient and his autologous unit in a definitively positive manner. Some system needs to be developed to positively track these units from collection through infusion. There is also the issue of giving group-matched platelets to people who routinely request them. Some physicians will only give Type A platelets to Type A individuals and they won't give Type O platelets to Type A individuals. How much are you letting your consumers drive inefficient blood practices? We do have special requests on occasion. For instance, we may need a lot of antigen-negative blood when we are transfusing either multiple alloimmunized patients or patients in whom we are trying to prevent multiple alloimmunization. This is another area where our blood center can help us a great deal by providing appropriate antigen-matched blood. Finally, I will make one last comment. When I go to the supermarket to buy milk, I consciously look for the milk with the longest shelf life. You wonder what happens in the transfusion services when staff members are ordering blood from the blood center. Is it the individual facility's benefit that is overriding, or is it the public or the greater benefit? I don't know, but my milk analogy makes me wonder. Questions/Comments Robert Travis: Just a curiosity question. Out of any given hundred days, how many days is your inventory at your optimal level? Arthur Bracey: Good question. I would say that we are at our optimum level about 60 percent of our days.

OCR for page 51
This page in the original is blank.

OCR for page 51
--> Logistic Problems and Perishables: The Kroger Company and Supermarket Seafood Robert Fields The Kroger Company is the largest grocery retailer in the United States. We have about 1,200 stores spread out over 20 different states all over the country. We buy and ship over one billion pounds of perishable product annually, and that perishable product has a shelf life from 3 days up to about 45 days for cheese and processed products. Thus, we have a tremendous jigsaw puzzle that we are putting together every day. I am responsible for all the procurement aspects of all the meat commodities and categories in our Kroger stores. I am also responsible for all the quality assurance procedures and also for following up on distribution for getting all the meat all over the country. There are a lot of similarities between what we do at Kroger and what you do with the blood supply. I will give you an example of how we are solving some of the logistic problems, some of the waste problems, and some of the distribution problems we have out there. In terms of compliance and safety issues, we deal with the USDA, the FDA, local and state agriculture departments, and local health departments. We have a series of procedures and guidelines that we have to follow. One of the problems that we deal with is that different health departments have different guidelines. We are constantly having to work under all the realms. Sometimes we even have instances in which the USDA or federal regulatory committee will have one set of rules, but a state or a local agency will knock those out and impose their own rules. Sometimes there is even some one-upsmanship or some competition. To combat that and to guarantee that we are in compliance with all the regulations and procedures, the Kroger Company has developed one of the most thorough quality assurance programs in the industry. We have quality assurance staffs in each of our marketing areas. We are constantly evaluating product quality coming into our distribution centers. We are also constantly going out to the stores and monitoring quality levels out there, monitoring temperature, shelf life, and all types of bacteria.

OCR for page 51
--> We also have a corporate staff in Cincinnati, which talks directly with Washington, D.C., and communicates all the new regulations out to the stores. I would like to concentrate today on our distribution and merchandising principles for the seafood department that might apply to your business. Much like your blood supply and blood usage, there isn't any demand pattern or supply pattern. Because seafood is an industry that still depends upon a hunted species, we can't count on our supply at any one point in time. Also, we never know what the customers are going to want in our stores. In many regards, then, our selling pattern varies widely, just like your blood supply. Price fluctuation is another comparison area. I understand that pricing is not level in your industry. Similarly, in seafood our prices can fluctuate 10 to 40 percent in a matter of days. In one or two days, it can go up a dollar or two dollars a pound. One of the challenges that we are looking at in seafood is the huge swing in supply and demand. Another key issue is temperature control. It is one of the most important areas in controlling quality and safety. We must keep our seafood products under 35°F, and the colder we can keep them, the longer the shelf life and the higher the quality we can maintain. Fresh seafood typically has seven days of shelf life from when it hits the docks. We have to be able to push that product through our entire system and still leave a couple days for the consumer's refrigerator, to ensure the quality level. Product cost varies. We may have set a retail price for sole in our stores of $6.99 a pound. If the market fluctuates the next day and our cost goes up, we have basically lost all of our margin. Thus, fluctuation is a big problem for us. In addition, we are trying to guess what the customer is going to buy. We have the same inventory problems that you have. We are putting seafood out in the case, not knowing what is going to sell that day. Our seafood shops are operating more or less as a convenience to the consumer. Our waste in our actual stores can sometimes hit 20 to 25 percent. That was a critical point that we had to address. How do we lower the waste in our stores? Our solution was a new type of distribution network. Distribution for us involves taking products from Canada, Mexico, the Gulf of Mexico, the Northeast, and South America and distributing them over 1,200 grocery stores in 20 states. That is a logistical nightmare. Until recently, what we were doing was a little bit like the Red Cross scenario. We had a hub, which was at our Greensburg Seafood Distribution Plant. We would bring all the products into that distribution facility, spread them out to different regional warehouses, and then once again put them on another truck and put them into the stores. We were cross-trucking product four or five times before it actually got to the store. We have been in discussions with people at Emery Air Freight, and they created a tremendous software and technology package for us. With it we are going to forward contract with our suppliers in different regions. We will

OCR for page 51
--> negotiate costs, product specs, and supplier or inventory deliveries. Once those negotiations are completed, Emery is going to tie in with our individual store ordering system via satellite. Each store is going to be able to scan a little bar code or a special number that is going to indicate, say, Dover sole. That order will be transmitted via satellite to Emery, which will basically be our hub, and then Emery will automatically throw that out to the suppliers. Those suppliers will get the product, Emery will arrange pickup, and that product will be delivered directly to our stores through Emery. The key advantage to this system is that we can order a product at 2 P.M. today and have guaranteed delivery of that product into our stores by 11:30 A.M. the next day. We have reduced four days worth of float on all the products we had going all over the system. That is inventory management. There is a lot of money saved on that, and we have added four days of shelf life to our product where it needs to be, right there in front of the consumer. In turn, we are going to reduce our waste. We think we are going to reduce our waste by half right off the bat, and hopefully knock it down to somewhere around 5 to 7 percent. Another big advantage is that we are able to give our consumers the variety they need instantly. If a customer comes in and needs a special request, it currently takes four to five days to get it. Now, it will only take 24 hours. Basically, that is what we are doing in the distribution network, but there are a couple of comments I wanted to make. Our customers are just like your donors in a lot of ways. One of the things that we found out is that to keep a customer will cost a dollar, but to bring a customer back costs you 10 dollars. Just as with your donors, it is very important to keep that customer because in the long run it is going to save money. Another thing that we demand of our stores is that they treat the customer like you would want to be treated yourself. That helps us keep that customer there. Questions/Comments Paul Russell: Do you ever get a product out in one of your stores and notice that it is not selling there, and think it ought to be in Cincinnati or somewhere else? Robert Fields: Not in the seafood scenario. With beef or some longer-shelf-life items, I might transfer them over to another division. What makes that possible is we have a centralized procurement department that coordinates all that effort out to all the different places. We'll be able to do it even better with our satellite system.

OCR for page 51
--> Peter Logue: When you do transfer that product, do you lower the cost to the customer, as it is now, I would assume, not as fresh? Robert Fields: We are operating all the stores under the same umbrella. It is the Kroger Company, and basically we pass on the same cost that that division bought it at to the next division. There are limits of course. We are not going to send a problem to another division and create another problem. Your first loss is your best loss, so if you are going to have to discount, you might as well do it in the division where the problem began. William Sherwood: It sounds like your transportation company, Emery, is going to manage the whole inventory for your seafood business to the point of buying and selling and shifting and making all the decisions because they have the computer system. Robert Fields: They have the technology. That is correct. The Kroger Company is going to continue to negotiate cost and product specifications, but Emery is going to turn into more or less an order placer, which will take the orders from my stores and distribute them out to different suppliers.