Bullwhip Effect In Healthcare Sector

In Supply Chain management, Bullwhip effect has attracted some considerable importance in the FMCG sector in the last 20 years. The phenomenon “Bullwhip Effect” shows how small changes at the customer’s end have tremendous impact on the operations back there in the chain. The cause behind this effect is mainly attributed to the increasing demand variability in the supply chain. We consider the number of companies taking part in a single supply chain. Each of the company gets the order from its immediate downstream company to be fulfilled. In Bullwhip effect the orders to the suppliers tend to have a larger variance than sales to the buyer. This demand distortion then gets amplified while propagating upstream.

The paper mainly focuses on how the Bullwhip effect the matching of the demand and supply and the main causes leading to this. It also focuses on the P&G case study to understand the Bullwhip effect and the steps taken to dampen the bullwhip effect. Further it aims at quantifying the bullwhip effect and reducing its impact on the supply chain.

Acknowledgement

I am thankful to Mukesh Patel School of Technology Management and Engineering (MPSTME) for giving me this opportunity to understand the Bullwhip Effect and the major causes and the P&G case. Furthermore helping us understand the various techniques and procedures for writing a research project.

I would also like to express my sincere gratitude to my college mentor Prof. Jinu Kurian for her continuous guidance and support and being available at all times whenever the need may arise. I would also like to thank Prof. Prasad Chakrabarty for his help and support in the project.

Table of Contents

Introduction

The bullwhip effect occurs when the demand order variability in the supply chain is amplified as they moved upstream in a supply chain. Distorted information from single end of a supply chain to the other may lead to tremendous inefficiencies. Companies can effectively deal with the bullwhip effect by thoroughly understanding its underlying causes. Organizational leaders are implementing various innovative strategies that pose new challenges:

integrating new information systems

defining new organizational inter relationships

implementing new incentive and measurement systems.

At macroscopic level, Bullwhip Effect generates inefficiencies in production scheduling, sourcing, capacity utilization, distribution and the profit generation. While if we take a look at the operating level it induces additional inventory which is placed inappropriately to maintain the service levels. It also reduces the performance level by decreasing the incoming cash and the potential revenue. It can even dilute any company’s competitive strategies.

2 factors have changed the landscape of supply chain management in the last few years significantly.

Availability and use of the technology and software applications to allow capture and sharing of information across a supply chain mostly through extranets

Increasing indulgence of members of the supply chain in order to move towards to put aside the traditional arms-length relationships amongst each other and in its place move towards closer, partnership-type arrangements. Again, the networks over which these collaborations take place must include the necessary levels of performance, scalability, security and reliability to realise these benefits to their maximum potential.

Problem Statement

To analyse the Bullwhip Effect in healthcare sector and to find out its causes, consequences and cures.

Significance

Global competition in the world market today contains many challenges to achieve a

degree of predictability in the supply chain and remove the impact of Bullwhip Effect.

Researchers have examined the bullwhip effect and some models to reduce it, however a very few research has been done on analysing the Bullwhip Effect in healthcare sector and to find out its causes, consequences and cures quantifying the effect of bullwhip and its

Measurement still remains an exigent research path.

Research Methods

A case-study based approach was employed to conduct the research. Data were gathered primarily through interviews, observations and archival sources. At the hospitals, Interviews were conducted, in person, with doctors, nurses, administrators, and other hospital staff. At the diagnostic laboratory, personal interviews were conducted with the CEO, pathologists and other lab technicians. As far as data gathering is concerned, we decided to employ an informal, minimally structured, non-directive interview approach, enabling to minimize the influence of the assumptions.

.

Literature Review

Fluctuations in demand have a varying graph when we compare from industry to industry. Driven by seasonal demand and business cycles, apparel industry faces major demand adjustments, while the diaper market is subjected to constant demand in the market arena. Due to misjudgement of demands, the big players in the retail market can be subjected to inventory shortages or surpluses. But given the consistency of demand in the diaper market the diaper supply chain should be more efficient and accurate. But it isn’t the case. The diaper market even with the reliable demand patterns isn’t able to match the demand-production matching. The major cause to this supply chain inefficiency can be subjected to “Bullwhip Effect”. The term was coined by Procter & Gamble who noticed amplification in the information distortion as the information of the order travelled upstream in the supply chain.

Bullwhip effect or Whiplash effect can be majorly seen in the forecast driven distribution channels. It indicates a lack of synchronization among the members across the supply chain. Even if there is a small fluctuation in the customer sales, it reflected upstream in an amplified form. Because of this supply patterns does not match the demand patterns resulting in inventory surplus at various stages of the supply chain.C:UsersIndiaDesktopblwhp.jpg

As the customer demand would be rarely perfectly stable, businesses should forecast demand in order to match the demands with the production and managing their inventory levels.

Some of the major reasons behind bullwhip effect are:-

Forecast Errors

Overreaction to backlogs

Lead time variability

No communication and no coordination along the supply chain members

Delay in information and material flow

Price fluctuations

Product promotions

Order batching

raw material orders from Procter&Gamble to its suppliers fluctuated over time. On further noticing it was found that farther down the chain, when sales at retail stores were studied, it was found that the fluctuation which was present, were small. It is reasonable to assume that the consumers of the diapers at the last stage of the supply chain used them at a steady rate. Although consumption at the end product was stable, orders for raw material were highly variable, increasing costs and making it difficult for supply to match demands.

Lack of coordination between supply chain members also results if information distortion occurs within the supply chain. Considering the Bullwhip effect P&G observed in the diaper supply chain. As a result of the bullwhip effect, orders P&G receives from its distributors are much more variable than the demand for the diapers at retailers. The lack of the supply chain coordination between members has an adverse effect on manufacturing cost. It increases the manufacturing cost in the supply chain. P&G and its suppliers must satisfy a stream of orders that is even more variable than customer demand. Procter&Gamble responded to the increased change by either building excess capacity or by holding excess inventory, both of which increase the manufacturing cost per unit produced. It even increases the replenishment lead time in the supply chain. The increased variability due to bullwhip effect makes scheduling at P&G and supplier plants much more difficult as compared to a situation with level demand. There are times when the capacity which is available and inventory cannot supply the orders coming in. This results in higher replenishment lead times in the supply chain from both P&G and its suppliers.

It even hurts the level of product availability and results in more stock outs in the supply chain. Very high fluctuations in orders make it difficult for P&G to supply all distributor and retailer orders on time. This increases the likelihood that retailers will run out of stock, resulting in lost sales for the supply chain. It also leads to an increase in the inventory costs. To handle the increased variability in demand P&G has to carry a higher level of inventory than would be required if the supply chain was coordinated. As a result, inventory costs in the supply chain increase. The high levels of inventory also increase the warehousing space required and thus the warehousing cost incurred.

It impacts the transportation cost in the supply chain. The transportation requirements over time at P&G and its suppliers fluctuate with the orders being filled. As a result of bullwhip effect, transportation requirements fluctuate significantly over time. This raises transportation cost because surplus transportation capacity needs to be maintained to cover high demand periods. It also leads to the increase in labour costs associated with shipping and receiving in the supply chain. Labour requirements for shipping at P&G and its suppliers fluctuate with orders. A similar fluctuation occurs for the labour requirements for receiving at distributors and retailers. The various stages have the option of carrying excess labour capacity in response to the fluctuation in orders. Either option increases total labour cost.

P&G estimated that due to the manual interventions required in their process of ordering, billing and shipment systems, each deal to its customers cost between $35 to $75 to process.

Sharing point-of-sale (POS) data across the supply chain can help reduce the bullwhip effect. A primary cause for information distortion is the fact that each level of the supply chain uses orders to forecast the future demand. Given that orders received by different levels vary, forecasts at different levels also vary. If retailers share POS data with other supply chain stages, all supply chain stages can forecast future demand based on customer demand. Sharing of POS data helps reduce information distortion because all stages now respond to the same change in the customer demand. Sharing aggregate Point Of Sale data is sufficient to reduceinformation distortion. P&G has convinced many retailers to share demand data. P&G in turn shares the data with its suppliers, improving coordination in the supply chain.

In a continuous replenishment programs (CRS), the wholesaler or the manufacturer replenishes the inventorey regularly based on Point of sale data. In its simplest form, CRS seeks to allow more accurate production planning of inventories and also matching of supply and demand. Success with continuous replenishment programs is achieved when production planning has become demand-driven on an end-to-end basis throughout the supply chain.

. Vendor-managed inventory (VMI) is a family of business models in which the buyer of a product provides certain information to a supplier of that product and the supplier takes full responsibility for maintaining an agreed inventory of the material, usually at the buyer’s consumption location (usually a store). A third-party logistics provider can also be involved to make sure that the buyer has the required level of inventory by adjusting the demand and supply gaps. P&G now employs vendor-managed inventory (VMI) in its supply chain, starting with its supplier, 3M, and its customer, Wal-Mart

In VMI, packaged-goods behemoth Procter & Gamble (P&G) initiated a “value pricing” scheme for sales to retailers. Value pricing was P&G label for everyday low pricing (EDLP), a pricing strategy under which retailers are charged a consistent price rather than a high baseline price punctuated by sporadic, deep discounts. P&G had many reasons for converting to EDLP. Its sales force had grown to rely on discounting to drive sales and the use of deep discounts had spiralled out of control, cutting into earnings.

Background

SMS Hospital, operating in Jaipur, is one of the largest government healthcare providers in Rajasthan. It consists of more than 500 beds and offers healthcare services ranging from a 24 hour accident and emergency service, to acute medical and surgical care, operating theatre suites, intensive care, ocus on cardiac and orthopedics areas and head injury and orthopaedic related trauma and community care. The hospital services more than 40,000 patient admissions and an average length of stay of 3.62 days for all patients.

There are10 operating rooms (OR) with state of the art equipment, in-patients have hospital wards, acute cases beyond normal post-operative hospital care have rehabilitation unit for recovery and related ancillary services in the hospital facilities. These rooms are being used for major inpatient procedures. On an average of about 27,000 operations are performed in these rooms in a year. Specialty and staffed with suitably qualified nurses and technicians organizes 500+ beds in the hospital. The Operating Room Management Department (ORMD) is the central division, which faces daily challenges of managing and allocating the staff and equipment so that surgeries can be performed in an efficient, cost effective, and safe manner. The equipment that is owned by the hospital may either be dedicated to a theatre or be shared across theatres. The Operating Room staff is comprised of registered nurses, nursing assistants, scrub technicians and unit secretaries and an Operating Room nurse supervises all these. Equipment to be used for Operating Rooms may either be owned by the hospital or be procured from outside on a loan/hire basis. The Operating Room Management Department is responsible for assignment of theatre suitable for the scheduled surgery and ensuring the availability of staff and equipment when needed.

Quality and cost of patient care implications are on overstaffed, undermanned, and unbalanced nursing teams. The hospital has adopted a hybrid strategy to meet its nurse requirements in both wards and Operating Rooms. The hospital has employed a group of permanent, full time nurse staff, which is also, complemented by a bank of casual, part time nurses which are available at short notice. Also, the hospital also has made some arrangements with several private agencies, which are known for providing qualified nurses for short periods to hospitals, on a temporary basis. Fifty % of the hospital’s requirements) in the terms of nurse requirements are usually been met by permanent staff, thirty% by part time nurses and twenty% from agency staff.

The hospital haven’t employ any specialist/surgeon, but only has a small group of physicians for overall supervision of the wards. Instead, it has arrangement with a large group of consultant surgeons who hold privileges to avail of the hospital facilities for treatment of their patients. The hospital derives its revenues from the fees charged for the use of its facilities – ORs, hospital beds and other services from the patients. In most cases hospital charges are borne by patients’ medical insurance providers and the hospital’s revenue is based on negotiated rates with these firms.

Value Chain

Figure 1 illustrates a schematic description of SMS’s service value chain. As shown in the figure, the multi-stage process begins with the patient’s consultation with his or her primary care physician. Upon receiving a referral from the general practitioner to visit a medical specialist, the patient will then be assessed by a specialist. This stage does not involve the hospital and may consist of several visits to the doctor’s office and diagnostic laboratories, deals with the assessment of patient’s condition and treatment options. A determination of the need for surgery results in the addition of the patient to the doctor’s list of patients requiring surgery/hospitalization. Typically, specialist doctors have privileges at several hospitals in the area, and the choice of hospital for surgery is determined by a number of factors that include cost, patient preference, case complexity, wait involved, facilities and other services provided by the hospital. Following the choice of hospital, the patient is scheduled for surgery by assigning a slot in one of the doctor’s theatre sessions at the hospital. It is not uncommon for elective surgeries to be scheduled several weeks to a few months in advance. The schedule for each session along with patient and surgery details is expected to be communicated by the doctor’s office to the hospital well in advance to enable the hospital to make adequate arrangements. The surgeon’s operating list is not frozen and is dynamic with additions possible at a later stage (in some cases a few hours in advance) and the hospital is expected to be responsive and be able to provide the requisite support of staff, equipment and supplies.

Surgery represents the third stage of the value chain and Saint Mary’s hospital is responsible for providing required support services. These include staffing the theatres with qualified nurses and technicians, and making available all equipment and supplies needed. The hospital is also responsible for pre-operative care of the patient and getting the patient ready for surgery and for post-operative care. Insufficient capacity and delays in these phases can result in blockage and starving of theatre resulting in underutilization of ORs with consequent adverse impact on hospital costs and efficiency. Post-operative care in hospital wards and in subsequent rehabilitation areas, if necessary represent the fourth and fifth stages before the patient is discharged and exits the

hospital system. Constraints on bed capacity in wards can lead to fewer surgeries being scheduled resulting in lower theatre utilization. This is in contrast to the rehabilitation services, where capacity shortage may only lead to patients being off-loaded to other facilities owned by competitors, thereby resulting in loss of potential revenue.

Planning and Scheduling at SMS Hospital

Planning and scheduling at Saint Mary’s hospital is similar to other hospitals in Australia with annual theatre plans forming the basis for hospital’s activities. The theatre plan involves assignment of theatre sessions to consultant surgeons and essentially defines the demand for hospital’s services. A half-day slot is considered the basic unit for this purpose and thus each theatre has a capacity of 10 sessions per week, based on a five day week. As described earlier, surgeons expect hospitals to provide complete flexibility in organizing the activities within their sessions and hence assignment of a theatre session to a surgeon commits the hospital for providing the necessary staff and equipment for operation and post-operative care, and hence the hospital workload is a direct consequence of the theatre plan.

In practice, theatre planning at Saint Mary’s begins three to four months in advance, in September/October for the following year. The plan is developed in consultation with the surgeons and varies little from year to year, except for changes to accommodate vacation periods of surgeons and other planned absences. The plan can be characterized as cyclic with effort to evenly spread out sessions assigned to individual surgeons. Further, to the extent feasible, weekly schedules are adopted. For example, a surgeon with 50 sessions would be assigned one session per week, usually in the same time slot every week (say, Monday morning). One feature of session plan at Saint Mary’s Hospital that merits special mention relates to the practice of concentration of sessions assigned to particular specialty. As a result, we find, sessions corresponding to different specialties peaking on different days of the week. Such practices are not unique to Saint Mary’s and are quite common at other hospitals in Australia.

The annual session plan forms the basis for nurse and technician staffing in the hospital. The labor cost of nurses and technicians is perhaps the most significant controllable operating cost at Saint Mary’s and directly impacts the hospital’s financial viability. Staffing plan for nurses in the OR theatres is prepared for each calendar month on a monthly basis, two weeks in advance to conform to regulatory requirements. The initial plan is developed on the basis of the session plan. The nurse schedule is revised and frozen only a day in advance, at which time the requirements are assessed based on surgery lists available in the hospital. Besides adjustments for providing the required number of qualified nurses, the final schedule takes into account deviations from the initial plan (for example, absence due to sickness, or inability to schedule the nurse due to excess overtime on the previous day etc.) At this stage there is only very limited flexibility in respect of permanent, full time staff and the nurse scheduler relies on part time staff, supplemented by supply from external agency to meet the requirements. As mentioned earlier, current practices at Saint Mary’s hospital result in 20-25% of nursing needs being met with agency sources. In monetary terms, the proportion is higher due to higher rates paid for the temporary staff. Nurse staffing in wards is similar and schedules are frozen only a day in advance.

Observations

Our interviews with doctors/surgeons indicated that in a majority of specialties there is no discernable pattern in the demand for elective surgeries. This is in contrast to illnesses such as flu, hay fever that exhibit pronounced seasonality. Thus, daily demand for elective surgery – end customer demand can be reasonably be described as fairly uniform with no seasonality (day of the week, month etc.) and low to moderate variability. Figure 2 describes the pattern of admissions (which corresponds to demand) for orthopaedic surgeries at Saint Mary’s hospital by day of the week. The figure is based on data for one year and excludes the holiday period (mid-December to mid-January) in Australia during which most doctors avoid scheduling elective surgeries. The data in Figure 2 shows mild seasonality by month (admissions range from a low of 332 to a peak of 512 with mean of 470) and high variability by day of the week. Excluding the weekends, the average daily admissions range from a low of 58 on Fridays to a peak of 123 on Wednesdays. The pattern is similar for other specialties, except for the location of the peak. While in Figure 2 the peak occurs on Wednesdays and Thursdays, for other specialties it might occur on other days of the week. Taken in conjunction with our premise that end customer demand is fairly constant, Figure 2 suggests strongly presence of phenomena similar to the bullwhip effect. In the remainder of this section we discuss briefly the impact of this pattern on hospital performance – operational and financial.

First, increased variability in the number of surgeries performed directly impacts the demand for post-operative services in the wards. Consequently, on peak days, shortage of beds makes the wards the bottleneck, thereby restricting the number of surgeries and reducing the theatre utilization and hospital throughput, which in turn leads to lower revenues and lower operating profits. While the hospital is a not-for-profit organization, operating profits represent one of the key funding sources for financing investments in equipment and facilities, and lower levels or absence of operating profit can severely restrict the hospital’s ability to provide state-of-the-art high quality service.

Second, the demand for nursing services – both in theatres and wards is directly affected by higher variability and results in higher labor costs. With Saint Mary’s strategy of meeting the nursing demand with a mix of full time, part time and agency staff, higher variability of demand translates to the need for higher levels of flexibility and larger proportion of casual and agency staff. Higher wage rates for these categories increases operating costs and leads to lower profits.

Third, SMS’s reliance on flexible (part time and agency) staff affects the hospital’s operating performance in more subtle, but significant manner. Higher levels of temporary staff leads to frequent changes in the composition of support staff assigned to each theatre, thereby inhibiting development of cohesive support teams. Consultant surgeons in our interviews mentioned this factor repeatedly as the primary cause of inefficiencies in theatre, requiring more operating time and reduction in efficient use of theatre capacity. The role of cohesive support teams in improving OR efficiencies is well known and has been noted in many other hospital environments. In the absence of such a support team, the surgery list scheduled for each session tends to be shorter (thus reducing hospital revenues). Alternately, with normal list the theatre time will be higher than planned, resulting in overtime and increasing operating costs. In either case, this practice adversely affects hospital’s profits. Furthermore, this factor influences the doctor’s choice of hospital, prompting use of Saint Mary’s for more complex cases requiring its facilities. For simpler cases, the doctors tend to choose other hospitals with more efficiently organized OR theatres, perhaps with more limited range of services. As the more complex cases typically involve higher costs, this behavior has an adverse impact on Saint Mary’s finances. Under the Australian system the fees payable (to the hospital and doctor) are based on predetermined rates for each class and do not depend on the case complexity within each class.

To summarize, increased variability in the demand for elective surgeries at Saint Mary’s hospital results in lower efficiencies, higher operating costs and lower revenues leading to lower profits. Thus, the presence of bullwhip effect leads to degradation of performance as in the case of manufactured goods reported in the literature. While this is intuitive and not surprising, it is interesting to note the dynamics are different from those observed in manufacturing – higher levels of inventories and shortages.

Macintosh HD:Users:vasvigakkhar:Desktop:Screen Shot 2012-12-29 at 2.09.56 AM.png

Macintosh HD:Users:vasvigakkhar:Desktop:Screen Shot 2012-12-29 at 2.13.36 AM.png

Methods for coping with the Bullwhip Effect

The preceding section highlighted the different causes of the bullwhip effect in the healthcare sector and identified strategies to mitigate the detrimental impact of the effect. We briefly summarize the strategies (see Table for a summary):

(i) Reducing uncertainty – In order to minimize or eliminate bullwhip effect, it is important to reduce uncertainty throughout the service value chain. If information about customer can be centrally managed and if each stage of the service value chain can be provided with latest information in a timely manner, it will help eliminate duplication of effort at different stages and also ensure data consistency across all groups that require that data. It is also critical that dynamic updating of information is performed which will greatly help reduce uncertainty.

(ii) Reducing variability – There is inherent variability in the customer demand process and any effort that can be taken to minimize this variability will help reduce the overall bullwhip effect. Also, localized practices that lead to increased variability in demand at any of the stages should be eliminated with better coordination. Use of good forecasting methods will also assist the reduction of variability.

(iii) Developing strategic partnerships – The bullwhip effect can be minimized by developing key strategic partnerships with different links in the service value chain. For example, in case of hospitals, the hospital administration has to forge strategic alliances with medical specialists. These partnerships will change the way in which the information is shared and will help foster better coordination between the two parties. Although this may not fully eliminate bullwhip effect, it will go a long way in reducing the effect.

(iv) Realignment of incentives – One prime reason for the lack of coordination and meaningful information sharing is that the different entities are evaluated on the basis of different criteria and receive rewards for different activities. The evaluation and reward system should be modified to stress cooperation across stages and so that the planning is performed on system-level objectives and not on division/local level objectives.

(v) Improving coordination across stages of service value chain – To reduce the bullwhip effect, it is also important to ensure coordination across stages that are facilitated and improved.

Summary and Future Research

In this paper we have identified and described a phenomenon in healthcare industry that is similar to the bullwhip effect observed in the manufacturing sector which results in amplification of demand variability upstream in the service chain. Not surprisingly, this distortion likewise leads to performance degradation. However, the similarity with the manufacturing sector does not extend much further since the causes and the impact are somewhat different.

For example, while bullwhip effect in manufactured goods results in higher levels of inventory and shortages, in hospitals it leads to lower levels of throughput, higher operating costs and longer patient waits. Hence, initiatives and strategies for minimizing the bullwhip effect and its impact require a different approach that addresses directly the root causes. Our study also indicates that the potential for performance improvements is quite significant. While our study focused on hospitals and other allied institutions in the health sector, we conjecture that a similar phenomenon may be present in other service industries. For example, billing practices in consumer utilities (electricity, gas etc) and credit card services with monthly cycles leads to variable unbalanced loads in certain segments. Similarly, scheduling practices in auto repair industry result in some variability amplification. Together this suggests that a more detailed study of the bullwhip phenomenon in services is warranted and may lead to order-of-magnitude performance improvements similar to those realized during the past two decades in the manufactured goods industries.