INGREDIENTS FOR INVENTORY MANAGEMENT TUNING PURCHASE AND SALES AT JAN DEKKER INTERNATIONAL

(1)

INGREDIENTS FOR INVENTORY

MANAGEMENT

TUNING PURCHASE AND SALES

AT JAN DEKKER INTERNATIONAL

26 February 2010

MAAIKE KARIANNE VEENENDAAL Student number: 1514490

Jan Dekker International

Plein 13 no. 1 1521 AP Wormerveer Supervisors: Mr J. Hermans Mr H. Botterhuis University of Groningen

Faculty of Management and Organization Landleven 5

(2)

ABSTRACT

Jan Dekker (JD) is a company supplying ingredients for the cosmetic and food market. Most items are traded and some are produced in house. JD gave impetus to the research to improve the offered performance to customers. The performed research is subdivided in an explorative research to investigate the problem and an in-depth research to design a possible improvement. The performance objectives are identified via interviews with staff who are in direct contact with the customers. To get insight in the current performance of JD, first the customer complaints are analyzed, because these reflect the non-satisfaction of customers. The main reasons for

complaints are late delivery (45%), Product non-conformity (15%), Labeling (12%) and Missing CoA (Certificate of Analysis, 12%). Second, the internally perceived performance is measured via a questionnaire. It investigates the relative importance of objectives and their perceived performance. The results are that Jan Dekker underperforms on customer requirements which are mostly related to delivery. Both analysis show the same result, namely that JD does not meet the requested level of performance on delivery.

The underlying causes of underperformance on delivery appear to be interrelated. They relate to purchase orders issued too late, lead times that are not respected, little insight in time-phased inventory levels, no insight in future sales and bad communication internally and to customers. The explorative research concludes by specifying three points for improvement:

• Inventory management system; • Forecast;

• Information and communication structure.

The in-depth research focuses on the design of the inventory management system. The main objective is to develop the out-line for an inventory management system that suits to the situation at Jan Dekker and improves the availability of items against acceptable costs, risks and lasting shelf life on products.

The central question is connected to the objective: How should decisions be made on the most important variables in the inventory management system?

The inventory management system for trade items and production items is set up separately, because the demand on traded items is independent and on produced items is dependent.

However, the set up for the design is performed in the same way. First the demand characteristics are analyzed, by distinguishing dependent and independent demand and by analyzing the demand pattern. Second, the inventory management system is selected based on the demand

characteristics and according to findings from relevant literature. Third, decisions are made regarding the most important decision variables, namely the order sizes, order moments and safety stocks.

Inventory management system for trade items: The demand on these items is independent.

(3)

system with fixed order points and fixed order quantities is supposed. The order quantity can be based on the minimization of cost in an EOQ, and the order moment depends on the demand during the lead time. A safety stock will buffer against demand fluctuations and hereby ensure a desired service level. This results in a minimization of total costs while guaranteeing service. However, this optimal situation holds out, because the shelf life on products works limiting to the maximum inventory level. To prevent the organization from high obsolescence costs, the maximum inventory should be limited to a relatively certain level of sales during the maximum shelf life period.

Inventory management system for production items: Items with dependent and independent

demand on them need to be managed with different inventory management techniques. It is assumed that the demand pattern of end items is comparable to trade items.

End items: Products which are frequently sold are made-to-stock, whereas infrequently sold items are produced make-to-order. The production capacity at Jan Dekker is limited, which makes that make-to-stock items are produced in maximum batch sizes. For these items a buffer stock is kept to protect against demand uncertainty. Make-to-order items are produced after receipt of the order in the volume requested by the customer.

Ingredients: The demand on ingredients can be seen as deterministic, which makes the demand levels predictable. Ingredients that are used for the production of items which are frequently produced should arrive 1 week earlier than requested. This improves the ability to pull production backward when there is a shortage of production capacity. The ingredients are purchased in such volumes that the total costs for ordering and holding are minimized. The Part Period Balancing method can be used to determine the cost minimum. Ingredients that are used for the production of make-to-order items are ordered up to a certain level. This level is equal to the requirements for one production run, so that the organization is able to produce the end item within a short term after receipt of a sales order.

When there is a combination of demand types, a mix of buffers should be kept. The order quantity is still based on limiting the order costs and holding costs.

This research has resulted in an inventory management system. It should help Jan Dekker to improve the availability of goods and consequently enable them to offer a better service to their customers. The system is designed in such a way that risk of obsolete goods and unnecessary costs for ordering and keeping inventory are limited.

(4)

PREFACE

Ten years ago I started my studies in Groningen and the theme of the introduction keiweek was “I came, I saw, I kei”, derived from “I came, I saw, I conquered”. According to Plutarch, the words by which Julias Ceasar succinctly described one of his victories. In Latin the words are “veni, vidi, vici.” Now, looking back on my period of studies and finishing my final thesis I would like to cite these words: “I came, I saw, I conquered”

I came to Groningen when I had just turned 18 to discover my new life in Groningen. Here, I started my studies in the field of International Technology. During my studies as well as during my social life, I saw, learned and discovered many things. My knowledge and personality grew of all experiences. After finishing my studies, I wanted to continue my studies in the field of my final project: logistics. Although I was convinced of the fact that I could do better, I had some difficulties with the level at the university. Before ending my studies I decided to leave Groningen, not to give up but to challenge an interesting project at Jan Dekker International. The problems that I faced at the organization were complex, which made the project complicated and sometimes like a struggle. Eventually, I conquered, and the proof is in front of you.

This final thesis is the result of a research conducted at Jan Dekker International in Wormerveer. It was a practical research in which theoretical knowledge of my studies has been applied to the situation of the company. The thesis was written during the final stage of my master Operations and Supply Chain Management, within the Business Administration department of the Groningen University.

When I read other researches, I dislike all the many expressions of thanks, but being myself at the end of this research I must admit that I faced many difficulties in which I appreciated the support of others. I realize that I could not have accomplished this result without the help of others and therefore I would like to thank everyone who provided me with information needed and advices. Especially, I would like to thank my supervisors at Jan Dekker for giving me the trust to perform the research and supporting me during my research. Thanks also to my colleagues, who have been very open and willing to provide me with their knowledge and information. Moreover, I would like to give attention to the people at the university, especially Mr. Land and Mr. Wijngaard, for their supervision, sharing ideas and other support during my research. Also my peer students need to be thanked for thinking along with me about the possible solutions for the problems I was facing during my research.

(5)

LIST OF ABBREVIATIONS

BOM Bill of material

BU Business unit

BUCI Business unit cosmetic ingredients BUFI Business unit food ingredients CoA Certificate of analysis

CSD Customer service department CV Coefficients of variation

EHSQ Environment, health, safety and quality department EOQ Economic order quantity

F&A Finance and administration department MM Material management

NCR Non conformity report PPB Part period balancing

PW Production and warehousing department SC Supply chain department

(8)

1 INTRODUCTION AND OUTLINE

This research is conducted as graduation project for the master Operations and Supply Chain Management at the University of Groningen. The research is performed at Jan Dekker International, a Dutch company located in Wormerveer. This chapter is the general introduction of the company and gives a general introduction to the first part of the research.

Jan Dekker International, founded in 1778, is originally Dutch and has a long history. The head quarters are still located in Wormerveer and by today it has six affiliates in Europe, employs 77 employees and supplies its products worldwide.

Jan Dekker international is a supplier of functional and active ingredients to the global cosmetic and food industries. The main businesses are cosmetic ingredients, featuring approximately 80% of their business. The products offered cover a brought product range, from bulk products like oil to high value ingredients like plant extracts. The other 20% are ingredients for food and feed application, with a strong focus on value adding ingredients. The total assortment exists out of approximately 600 products. The product range of Jan Dekker features following product groups:

1 Active ingredients 2 Oils and fats 3 Antioxidants

4 Preservative systems

Jan Dekker is mainly a trade company, which means that products are sold in an identical state as they were bought. The company is an exclusive distributor in Europe for some product ranges of some of their suppliers, which distinguishes them from competitors. Next to the directly traded products, there are also products which are physically not changed, but repacked into a different pack size, so that their customers are enabled to purchase smaller quantities. Finally, a small part of the product range is blended by Jan Dekker. These blends are mainly developed in own laboratories. The production is generally an uncomplicated process of blending ingredients in a kettle to a homogeneous product. The production is performed both internally and contracted out. The main tasks of management and distribution are centralized in the Dutch office; management, purchasing, warehousing and production are located here. Next to the central warehouse Jan Dekker uses decentralized warehouses at their affiliates to supply their local markets.

(9)

According to Slack & Lewis (2002) organizations can improve their competitiveness when they align the organization with its external environment. This means that an organization should organize its resources and process in such a way that it is capable to serve the customers requirements’. The performance experienced by the customer depends on the competitive factors valued. The customers require low costs and high performance on quality, delivery and flexibility. If an organization has good understanding of the requirements of the market, then the organization should be aligned to it in order to realize a fit between the requirements and the offered service and products.

This research exist of two parts; In this research first an explorative research will be performed to investigate current situation, described in Chapters 2 and 3, and in the second part an in-dept research is performed that will investigate the main point for improvement, described in chapters 4, 5 and 6. The aim of the explorative research is to investigate the current fits and gaps in order to describe areas for improvement. The first stage is the situational analysis in which the company will be described by its products, organization and processes, and the market by its characteristics and requirements. This is investigated by having semi structured interviews with employees who have customer contact; these are the business unit managers, sales employees, supply chain manager, and employees within supply chain. In the second stage the fit-gap analysis is performed in order to identify the areas in which Jan Dekker does not meet the required performance. This will be done by investigating customer complaints, because these directly reflect dissatisfaction of customers, and by a questionnaire among the interviewed employees about the internal perceived performance of Jan Dekker on the performance objectives. This way the performance gaps will be described and possible improvements will be proposed. This is concluded with a decision about the major gap to be improved.

(10)

2 ANALYSIS OF CURRENT SITUATION

This chapter gives a general overview of the current situation of Jan Dekker. It gives attention to both the internal and external situation. First the product characteristics are described, followed by the organization structure and the main internal processes. Second the external situation is described. Here attention is given to the market characteristics. The chapter is concluded with the performance objectives that describe the values for the market. This chapter is the basis for fit gap analysis in chapter 3.

2.1 Products

Jan Dekker supplies its products to cosmetics and food producers. The total product range of Jan Dekker features active ingredients, oils and fats, antioxidants and preservative systems. The organization is mainly specialized in ingredients which add high value to the end products. These are for example ingredients which capture special effects that provide the end products with certain desired characteristics. These products are aimed for the cosmetic and food market. Jan Dekker offers approximately 600 products which are sold in different pack sizes or stock keeping units (SKU). In total there are about 2400 different SKU’s.

All products have a limited shelf life, varying between 6 and 36 months. Due to its perishable characteristics, products should be sold before its expiry date to prevent costs for obsolete stock. The packaging of products must prevent the product from pollution and contamination. Moreover there are requirements to the storing of products that result from product characteristics or legislation. The storage of these products is partially outsourced to specialized subcontractors with for example cooled warehouses or safety depots. Legislation also places requirements on production, warehousing and transport processes.

Products are replenished by purchases or production. Most of the end products, approximately 80% are replenished by purchases. The lead times incurred with purchases vary between 1 week and 4 months. The long lead times mainly result from transport times from suppliers that are located in the Far East. Normally, the production process itself only takes a few hours. However, the end products are only available for sales after the quality is approved and this process can take up to 3 weeks. Most of the analyses are performed in the internal lab, but some highly specialized analyses are outsourced.

2.2 Organization

(11)

Figure 2.1: Organization chart

Supply Chain (SC) is operationally responsible for supplying customers on time with the products demanded. The department is divided in 4 sub departments that manage the separate steps of fulfilling orders. Customer Service Department (CSD) is responsible for planning delivery dates of sales orders and releasing them when stock is available. Material Management (MM) takes care of maintaining inventory levels by generating call offs from contracts and making the production planning. Production and Warehousing (PW) takes care of the production and order handling in the warehouse like picking and shipping of orders. At this department the sales and operational purchasing of the BU’s come more or less together, because the sales order are supplied with the products from the purchase contracts.

Finance and Administration (F&A) and Environment, Health, Safety and Quality (EHSQ) are both strongly supporting departments that are not of a major role in the sales and supply process. F&A is mainly involved in administrative tasks like invoicing and settling bills. EHSQ is an independent department with a highly monitoring role. They are responsible for controlling the quality of products and internal processes.

The division of tasks between the departments is very clear in the Dutch affiliate. The offices abroad are mainly sales affiliates from BUCI and BUFI and in addition the run certain supply chain tasks. The responsibilities differ per affiliate, and a more detailed description can be found in appendix 2.

2.3 Processes

In this paragraph the core processes that serve the customers are described; these are the processes of fulfilling sales orders, maintaining inventory so that sales orders can be supplied, and the value adding processes.

2.3.1 Sales and supply

Order processing of both BU’s is performed in the same way. Sales orders come in at the sales responsible of the BU, who negotiates the price and delivery conditions with the customer. After this is agreed on, the sales order is forwarded to the SC that is responsible for the order fulfillment. CSD enters the order in the ERP system Navision and confirms the possible delivery date to the customer. Products that are on stock can be supplied within few days and can directly be released, or else products can be released to ship after availability. Released orders are picked one day before shipment. The release of CSD triggers MM to create a pick order and prepare the shipping documents. The pick orders are sent to PW, where the orders get picked and provided with the shipping documents, labels and certificates of analysis. The next day the orders are

Managing Board

Business Unit Cosmetic Ingredients

(BUCI)

Business Unit Food

(12)

2.3.2 Inventory management and replenishment

Inventory is hold in order to accurately meet customer demand. However, some products are not kept on stock because these are not frequently sold. Products that are kept on stock are called stock items, and products that are ordered to fulfill a sales order are called order products.

The department MM is responsible for maintaining inventory to fulfill the orders, but the desired level of inventory is decided on by the BU manager. The desired stock level relates to the economic stock, which should be equal to approximately 2 months of sales. The level of 2 months was set, because most products can be replenished within 2 months. The BU manager sets the level of the desired stock level, but this does not appear to be related to a serious form of forecast or the annual forecast.

MM uses the desired stock level to decide on replenishing orders. When the available stock plus purchases minus the sales orders placed, is lower than the desired stock level, then a new purchase order should be placed. For the calculation, see also figure 2.2. There are no guidelines for order sizes, and they are mainly based on the experience of the purchaser or historic order sizes.

Figure 2.2: Replenishment to meet desired stock

+

Stock in the warehouse Purchase orders =

-

Available for sales Sales orders

= Economic stock *

* When the Economic stock is smaller than the desired stock level, then additional purchases should be made

The method used is very basic; a fixed order level and an undefined order size. It was chosen because it was believed to lead to good availability of products and to make it understandable for everyone in the organization. However, the method does not always lead to a good availability of products. One reason is that the desired levels are not regularly updated, so that sales growth is reacted on too late. Basically, the inventory management system is not based on relevant variables like lead times, demand variability and partially demand levels. As lead times are not taken into account, some products are ordered too late and others too early, causing inconvenient stock levels. Safety stocks are not consciously chosen, so that demand variety is not buffered. Also order quantities are not matched with the demand level, so that unneeded costs for ordering or holding are made. Altogether, products with a long lead time tend to stock out more easily, while the inventory level of products with a shorter lead time tends to be reasonably high and unneeded costs are made.

2.3.3 Adding value

(13)

Transforming processes

The number of transforming processes is limited, because Jan Dekker is mainly a trade company. The most obvious value adder is the production of products. Products are blended in a composition which creates characteristics which are desired in end products. Also repacking products in smaller pack sizes is of value for customers, because this way they are not forced to purchase big volumes which they do not need or do not prefer to open at once.

Services

The first two service values are related to logistics. Having inventory available on limited distance and in limited time is one of the most important values; it keeps customers flexible to place orders last minute and reduces uncertainties about lead times. For European customers the reduction of organization and paperwork for import is another value, as many products have an origin in the Far East.

Jan Dekker does also supply much service in terms of product information and development. Jan Dekker performs extensive quality analysis for most products and supplies analysis with every product to ensure the quality of the product. Second, new products are spotted in the market and are developed internally to offer to the customer, this way much knowledge can be transferred to the customers. Finally also customer oriented solutions are developed in cooperation with the customers.

2.4 Market characteristics

Jan Dekker supplies its products to two different markets, namely the cosmetic market and the food market. Both markets have specific characteristics and will therefore value particular performance objectives. The markets’ characteristics are described below.

Cosmetics

The cosmetics market is traditionally a luxury branch, where a high product quality focus is set. A current trend is that also price fighters become more active, which brings a pressure on prices of end products and ingredients. This pressure mainly applies to the bulk products, because for the specials there is less supply and customers are more loyal to their supplier. In this branch many product innovations are launched to the consumer market, which requires new formulas and ingredients. A current trend is that the producers are focused on shortening the number of suppliers, products and stocks and they constantly change order quantities.

(14)

important. Although customers attach value to different competitive factors, some main factors can be distinguished when looking at the main markets regions of Jan Dekker. The competitive factors valued by all cosmetics customers are in time supply of materials and information, and meeting quality and legal requirement.

Food

The fundamental difference between food and cosmetics is that food is a basic need of mankind. In general, there is a stronger cost focus in the food market and margins are smaller. Jan Dekker focuses on a market segment of high value products. For these products it is highly important to meet the quality specifications, and the pressure on prices is less tense.

In the food market there are less regional differences. Customers find it very important to be quickly supplied with information. When they want to know anything about the delivery of an order, they require direct and reliable information. When they ask a question for example about product characteristics, they want to be kept informed on the status of their request. When Jan Dekker is able to solve customers’ problems fast and with providence of enough information, this will have a great influence on winning the order. Reliable and timely information about supplies are essential to bind customers and else they will switch to a competitor. Moreover, the flexibility of the product pack size is important to many customers, because they want to be able to order in smaller quantities also.

In the next paragraph the market requirements will be used as a starting point to define the performance objectives for Jan Dekker.

2.5 Performance objectives

Different markets have different customer needs and therefore the performance objectives they value can differ. Although the market characteristics of cosmetics and food differ, they do value comparable performance objectives. Competitive factors are dimensions on which a company tries to compete. They describe the characteristics of a product or service that a customer can see or experience and attach value to. Competitive factors can be clustered in five common performance objectives: Quality, speed, dependability, flexibility and speed. The purpose of describing the competitive factors is to articulate market requirements in such a way that will be useful to operations. (Slack, 2002). Describing the market characteristics in this research serves as the starting point to define the performance that JD should achieve on the performance objectives. In chapter 3, the required performance by customers will be compared to the current situation within Jan Dekker in order to identify the improvements that should be made.

(15)

order fulfillment process in which they may express different values. It is remarkable that there are hardly differences distinguishable between the requirements of both markets and therefore all performance objectives for the markets are described and typified together.

Quality

Quality is one of the most important performance objectives and is order qualifying for these products. Quality of the products and additional reliable information and documentation is of importance because the ingredients they use do directly influence the quality of their end products and some are subject to strong legislation. Customer requirements related to quality are:

1 Purity of the products;

2 Orders delivered without damages or other defects; 3 Remaining shelf life;

4 Accuracy of information; 5 Accuracy of documents; 6 Usefulness of technical advice; 7 Transport safety;

8 Products are produced and processed conform ISO 9001/14002, Kosher, GMP, Feed, etc. Delivery (Speed and dependability)

The importance of delivery is that the availability of ingredients influences the planning of the customer production planning. Even when products are late, it is important to know as soon as possible if planning needs to be adapted. For inquiries about product applications and solutions, customers want to know that their request is being processed. The customer requirements related to delivery have a winning influence on orders, among them are:

9 On time in full delivery;

10 Lead time from order to delivery;

11 Reliability of initial promised date of delivery;

12 Timely information about the order state and possible delays; 13 Response time on emails and phone calls;

14 Correctness of place of delivery.

Flexibility

Customers require an increased (product) flexibility of suppliers. This results from their aim to reduce the number of suppliers, which in return have to be able to supply a wide range of products and packages. Flexibility is considered to be a winning factor. Customer requirements related to flexibility are:

15 Product range;

16 Assortment of packaging volumes; 17 Rate of new product introductions;

18 Think along with customers to develop customized solutions. Cost

(16)

This chapter functions as an introduction to the fit-gap analysis. In the fit-gap analysis it will be investigated to which extent the organization is capable to offer a good performance on the requirements of the market. In other words, Jan Dekker should provide a minimum required level at the performance objectives quality and costs, and should excel in the objective delivery and flexibility to distinction itself on the cosmetic and food market.

Conclusion

(17)

3 FIT-GAP ANALYSIS

This chapter aims to identify the gaps which will be driving for improvements during the successive in-depth research. A gap is the difference between customers’ view of what is important and the way in which the operation actually performs. (Slack & Lewis, 2002) Not achieving customers’ satisfaction can happen roughly at two moments. The first moment happens when Jan Dekker knows in advance that it is not able to meet the specifications in the customer request; this says something about the capabilities of the organization. The customer can then decide on not ordering at Jan Dekker, or to change the requests’ specifications. The second moment is when a customer request is confirmed, but for some reason the company is not capable to meet it; this says something about the reliability of the processes. This is visualized in figure 3.1.

Figure 3.1: Ability to meet customer request

The reliability of processes can be measured according to a number of measures, but it is mostly important to be good at the performance objectives required; described in chapter 2. The performance objectives have connection to both capabilities and reliability. In the fit-gap analysis it will be investigated to which extent Jan Dekker offers the requested level of performance. Two methods will be used; first a closer look is taken at internal registration on customer complaint registration, which will reflect the areas of dissatisfaction of customers. Second, a questionnaire is filled in by employees and their opinion reflects the performance of Jan Dekker perceived internally. The outcome of the analysis will be the starting point to appoint the most important gaps to be improved.

3.1 Customer complaints

Customer complaints are a statement of the market about non satisfaction about the products or services received. It gives the organization insight in the requirements of their customers and in the areas to be improved. At Jan Dekker complaints are administrated in a Non Conformity Report (NCR) system. The NCRs from 1st_{April 2006 up to 31}st_{May 2006 were analyzed. NCRs}

dated before April 2006 did not seem to be representative, because a new MRP system was implemented at the beginning of the year and might have affected the customer satisfaction. The

(18)

time. In total 65 NCRs were analyzed. To get insight in the reason and cause of complaints, the NCRs were categorized by the type of complaint and the responsible parties that had caused them. Appendix 3 gives an overview of both categorizations.

First we take a closer look at the reason of complaints of the total of NCRs. It becomes clear that the main reason for complaint is late delivery (45%). If we take a look at the parties that are responsible for the complaints on late delivery, then the transport company and Jan Dekker itself are mainly responsible for the inconvenience. Other prominent arias for concern are Product non-conformity (15%), Labeling (12%) and Missing CoA (Certificate of Analysis, 12%), illustrated in figure 3.2.

Figure 3.2: Causes of all NCRs (April-May 2006)

Communication; 3% Transportation; 2% Labeling; 12% Product non-conformity; 15% Late delivery; 45% Missing CoA; 12% Damages; 6% Stock; 2% Packaging; 3%

Second, we go into detail about the responsible parties that had caused the NCRs, illustrated in figure 3.3. For a number of them there were multiple possible parties and for these compounded groups are used. In the single groups, JD is the main responsible for NCRs (35%), followed by dissatisfaction caused during the transport. Also the NCRs caused by either one of them covers 12%. This means that together they are responsible for at least 73% of the NCRs, and appear to be the biggest area for improvement.

(19)

research, and therefore we concentrate on improvements that can be made internally at Jan Dekker.

Figure 3.3: Parties responsible for causing the NCR (April-May 2006)

JD or supplier; 5% Subcontractor; 3% Supplier ; 15% Customer or transport; 2% Supplier or transport; 2% Transport; 26% JD or transport; 12% JD; 35%

(20)

Figure 3.4: Causes of NCRs for which JD was partly responsible (April-May 2006) Communication; 3% Damages; 3% Stock; 3% Missing CoA; 15%

Labeling; 18% _Product

non-conformity; 24% Late delivery; 35%

3.2 Internally perceived performance

In chapter 2 the performance objectives were described. The performance objectives were derived by interviews, but it did not reflect the relative importance of them. To get insight in that, a questionnaire was sent out to the same people as interviewed; sales managers of both BU’s, the general director, SC manager and CSD manager. It inquires about the relative importance of each performance objective (low, medium, high importance), and about their perceived performance in the current situation (low, good, too high performance). The responses of the participants were valued and an average was determined. Even though this method is not suitable to draw hard conclusions, it gives a certain insight in the fields that are underperforming and performing well. This paragraph presents the performance of the organization on the objectives in a Platts Gregory diagram, figure 3.5.

(21)

Figure 3.5: Platts Gregory diagram

Quality

The performance objective Quality underperforms on deliveries with damages and on accuracy of documents. Damages are mainly caused during the transport. Accuracy of documents has to do with the documents that are supplied together with the delivery. Problems occur due to the fact that the products and related documents are not treated with enough care; as they get lost during handling and transport. As mentioned in the previous paragraph, Jan Dekker already started an improvement program with the forwarding agents in order to reduce dissatisfaction on these points.

Delivery

The other performance objective that is to be improved is delivery. The underperformances are interrelated issues about accurate information on new arrivals and stock levels on the sales site, accurate information about sales in the purchasing site and overall awareness of lead times. At this moment, purchasing has a bad insight in the approximate future sales, thus resulting frequently in too low availability of stock. New arrivals are planned tightly, internal lead times are often not respected and deliveries are planned tightly as well. This results in longer lead times, so that the initial planning cannot be met and products get supplied late. For Jan Dekker it is very important to outperform on delivery, because it is an order-winning factor. If they want to gain competitive advantage, which should result in a growing business, this performance objective should be worked on.

1 2 3 4 5

Quality Purity of the products X / O

Orders delivered without damages or other defects O X

Remaining shelf life X / O

Accuracy of information X / O

Accuracy of documents O X

Usefulness of technical advice X / O

Transport safety X / O

Products are produced and processed conform ISO etc X / O

Delivery On time in full delivery O X

Lead time from order to delivery O X

Reliability of initial promised date of delivery O X

Timely information about the order state and possible delays O X

Response time on emails and phone calls X / O

Correctness of place of delivery X / O

Flexibility Product range X / O

Assortment of packaging volumes X / O

Rate of new product introductions X / O

Think along with customers to develop customized solutions X O

Cost Price of the products X / O

(22)

3.3 Causes of underperformance and improvements

In this final paragraph of the explorative research the aim is to make a link between underperformance and the underlying reasons for it, in order to determine the most important areas for improvement. The second part of the research will investigate how Jan Dekker’s performance can be improved in such a way that the organization will be better capable to meet the requirements of the customers.

In the analysis of NCRs it appears that late delivery of orders is the main cause for non satisfied customers. The analysis of the internal perspective on the required and offered level of performance also shows that the delivery is the main concern of underperformance. Second field underperformance according to the internal perspective is the quality; this matches with second largest field of complaints, the Product non-conformity and also with the damages. In a reflection, it is good to notice that the internal perspective on what is important for customers does match with the dissatisfactions expressed by customers in the NCRs. There are two explanations for this; the most obvious conclusion would be that the organization has a good idea about the demands of the market. Contrary, it could also be that the organization is intimidated by the NCRs and looses sight of other well performing performance objectives. However, it seems to be mostly reasonable that in the first place delivery, but also quality, are most important to be improved. It seems that the underperformance on delivery relate to a number of processes in the organization that can be improved.

The first problem is that the replenishment method used by MM does not lead to a reliable availability of products. As described in paragraph 2.3.2, the economic stock level should be equal to two months of sales. This leads to inventory levels that do not match the sales. The economic stock at the reorder moment should be sufficient to fulfill the demand during the lead time, including the fluctuations. For products that have a reasonably stable demand pattern and the lead time is less than 2 months it should be possible to guarantee a reasonable good availability of products. Nevertheless, some products have lead times longer than two months and fluctuating sales, which nearly automatically results in stock outs. Another effect is the financial impact; products with a short lead time do always have a high physical stock, leading to higher costs. Moreover, the level of economic stock is not regularly updated, so that the organization always runs behind sales growths. To improve the match between purchases, available stock levels and sales, the first point for improvement is:

! Inventory replenishment system

The second problem is that MM has little insight in future sales. In the first place does the level of two months sales does often deviate from real sales. Also, this figure does not indicate for which period sales will be on this level. Having a bad insight in future sales makes it more risky to purchase material for a longer time period, because it is uncertain how long the products will remain in the warehouse and the risk of obsolescence increases. The second possible point for improvement is:

! Forecast

(23)

be supplied soonest possible, MM sometimes plans too early arrival dates, next the delivery dates are shortest possible planned after this arrival date. When delays occur on the arrival, MM does not always inform CSD about it or CSD does not use the information to inform the customer. Moreover, the analysis time that counts for some products is often not computed in the lead time, because it is not clearly set who is responsible for this, neither how long the lead times are. Improving the reliability of internal information, so that customers can be informed correctly, is the third point for improvement:

! Improving the transparency of information and the communication structure

(24)

4 RESEARCH DESIGN

The aim of the explorative research was to identify the areas on which Jan Dekker is not providing the requested performance to customers. It turned out that Jan Dekker did not match the requested performance level on all performance objectives. Having a too low performance on performance objectives will affect a companies’ competitiveness; customers will search for alternatives that do satisfy their requirements. The underperformance of Jan Dekker gives reason for the in-depth research. The problem statement for this research is:

Give the initial impetus to the realization of a performance level on performance objectives according to the requirements of customers in order to realize a competitive advantage.

The problem statement states that the level of performance on performance objectives should match the requirements of the market. It is therefore important that under performing areas should be improved. The realization of an inventory management system should enable the organization to improve the performance on delivery.

Currently the ERP system Navision is used, but the inventory management module has not been implemented. The control is performed manually and little tools are used to perform the task, which results in practice in late release of purchase orders, so that material arrives too late to fulfill sales orders in time. A simple and quick solution to all these points would be to increase the stock level. However, this action has negative side effects on the financial performance of Jan Dekker and also on the customer requirements on minimum remaining shelf life. Increasing the stock level will result in increases in working capital, holdings costs, depreciations on obsolete stock, and risks on non-saleable products, which all will indirectly also causes higher product costs for the customers. The in depth research will focus on the development of a consistent inventory management system that considers the most important relevant forces.

The main objective is to develop the out-line for an inventory management system that suits to the situation at Jan Dekker and improves the availability of items against acceptable costs, risks and lasting shelf life on products.

The central question of the in-depth research is connected to the objective:

How should decisions be made on the most important variables in the inventory management system?

The main question will be answered by giving answer to three questions, subdivided in some sub questions.

(25)

to be answered in the both chapters are the same; therefore the lay-out of the chapters is comparable, starting with typifying the demand.

How can demand characteristics be typified that are relevant for the inventory decisions?

The characteristics of demand are important for the inventory decisions, therefore the demand pattern needs to be analyzed. Two parameters are especially important.

1. Which dependent and independent demand items can be identified?

2. Do the items show a rather continuous or more discontinuous demand pattern?

To give answer to these two questions, the customer demand during the period from 01-01-2006 till 01-09-2006 will be analyzed. The sales order during this period will be leading to give indication about the demand characteristics. Therefore demand of individual items will be looked at in time periods. The analysis of the demand patterns is performed in chapter 5.1 for trade items and 6.1 for production items. After the demand pattern is defined, it is possible to decide on very basal inventory systems.

Which inventory management system is suitable for the different types of demand?

The inventory management system needs to be selected based on the demand characteristics. Jan Dekker is trading and producing items, which mean that there is independent demand on end items and dependent demand on ingredients. It is important to stress the difference between the two types of demand, because dependent demand can be determined quasi deterministically, while independent demand is uncertain. In theory different management systems are described for dependent and independent demand. A general outline for the inventory management system will be discussed on the basis of the relevant theories and practice at Jan Dekker in chapter 5.2 and 6.2.

How should decisions be made on the most important decision points in the system?

In the interpretation of the system there are three major decisions that are more or less related. The decision variable within any inventory management systems are:

1. What would be suitable order sizes?

2. When should replenishment orders be placed?

3. What would be a suitable safety stock to manage uncertainties in the system?

These questions will be answered with the support literature; Schönsleben (2004), Vollmann, Berry and Whybark (1997), Wild (1997), Waters (1992), Hax and Candea (1984), Tersine (1982). The variables of the inventory management system needed to be set; these are the order quantities and moments of the moving stock, and the safety stock level. How this is decided on is described in the chapter itself. This results in the basis of the inventory management system, described in chapter 5.3 and 6.3. After, it needs to be checked if the constraints result in any practical limitations; at this points the inventory management system needs to be adapted. The consequences of the constraints are described in chapter 5.4 and 6.4.

(26)

5 INVENTORY SYSTEM FOR TRADE ITEMS

Analysis and design

This chapter describes the characteristics of the trade products, the requirements on the inventory management system and the restrictions. Considering these three, a suitable inventory management system will be developed.

5.1 Characteristics of demand

Trade products are sold in an identical state as they are purchased, so they are not transformed. This means that all demand on these products is independent. When demand is dependent, the demand can be derived from demand for other products. (Schönsleben, 2004) Due to the independence of the demand, it is more difficult to determine the moment of demand.

At Jan Dekker it is believed that the demand pattern of many products is strongly fluctuating, however analysis of the figures moderates this assumption. To investigate the demand fluctuations two different analyses on sales data from 1 January 2006 to 1 October 2006 were performed. The starting point for the analysis was a document with all sales order lines in this time span. The document did not contain information on if the items were trade items or production items and for each item this information was manually looked up in Navision and added to the document. Also information about if the items were stock or order items had to be looked up in a separate spreadsheet. Now the trade products had to be distinguished from the ingredients and production items and the order items were filtered out. This resulted in a selection of 125 items in total. All sales orders were sorted per item and divided in time buckets of one month. The first analysis compares the total sales in number of SKU’s sold per month, and are translated to percentages of the total sales. This shows fairly constant sales throughout the year. (See appendix 4.1) Second, the coefficients of variation (CV) per SKU were calculated by dividing the standard deviation by the mean. The CV shows the dispersion of sales of an SKU around the mean. The results show that the CV varies between 0.30 and 3.0; 60% is smaller than 1. (See appendix 4.2) This means that the demand pattern is neither stable nor peaky demand, but is reasonably variable. Concluding, the demand of trade products displays certain variation that can be defined reasonably variable.

5.2 Selection of the inventory management system

(27)

Figure 5.1: Four types of inventory management systems (source: Silver, Pyke and Peterson (1998))

Order quantity Order up to level Order point (s,Q)- system (s,S)- system

Periodic review system (R,s,S)- system (R,S)- system

Each of these four systems has its own advantages and disadvantages. The selection of a suitable system depends on the characteristics that count. Some authors favor to use different types of systems for different types of inventory items. In that case, the ABC classification is often used to determine the importance of items, and suggested to select the inventory system. More information about the ABC classifications can be found in appendix 5. In the case of Jan Dekker, the understanding of inventory systems is not too high and the number of articles involved is limited with 125 SKU’s, which makes it recommendable to use one methodology for all products in order to protect from a jungle of control systems.

The first question to answer is: When should a replenishment order be placed? Hax and Candea (1984) state that periodic review systems are used when the stock status is not always known and it is therefore not possible to review the inventory status or when review costs are very high and do not counterbalance the costs of holding extra inventory. The advantage of continuous review systems is that less inventory is required for an item to provide the same level of service. At Jan Dekker, there are no restrictions to the moments of replenishment, besides the aim is to improve the customer service level while reducing the cost. Both arguments advocate a continuous review system at Jan Dekker. The only thing that is not looked at is the advantage of ordering products together at the same moment of time. However, this could easily be overcome by checking the stock situation for products of the same supplier when one of their products needs to be ordered. The second question is: How large should replenishment orders be? In situations where the unit-per demand transaction is highly variable the (s,S)-system is preferred, while fixed order quantities are more suitable for products with a less fluctuating demand. In the first paragraph of this chapter, the demand pattern was analyzed, showing a low variation overall. This advocates for the use of a fixed order quantity.

Concluding, at Jan Dekker it would be recommendable to use an order point in combination with fixed order quantities, this system is called an (s,Q)-system. In the following paragraph, the order point and order quantity will be determined.

5.3 Design variables of inventory management system

(28)

5.3.1 Order quantities

The economical order quantity is calculated by the formula for EOQ. The variable costs in this formula are expressed as a percentage of the cost price.

EOQ = √ [(2 * ordering costs * annual demand) / (interest rate * unit cost)]

An outline of the ordering costs and interest rate is given in appendix 6, which are respectively 150 Euros per purchase order and 18% per Euro per year. The interest rate is expressed in a percentage, determined by the total fixed costs divided by the average inventory value times the cost price of an item. For product A, the annual demand is 437 units and cost price 162 Euros per unit, both shown in appendix 7. The EOQ for product A results in 67.08 units.

The order quantity is not a discrete number of units. To determine the discrete value of the order quantity, the costs of higher and lower round offs are calculated and the quantity with the lowest costs is set. In the case of product A, the total costs are lowest at 68 units, see appendix 7.

5.3.2 Order moments

Ordering should take place when the available inventory is equal to the demand during the lead time, so that the demand can be fulfilled without stocking out. When demand is constant is equal to the demand rate, but in case of uncertainty in demand a safety stock is added. Safety stocks will be discussed in paragraph 5.4. The order level is a fixed level of inventory that can be calculated by:

Order level = lead time * daily demand (+ safety stock)

The lead time is available in days in Navision and to be calculated based on 252 annual working days. The daily demand can be deducted from the annual demand. In the case of product A, the lead time is 14 days, resulting in an order level of 24. The determination of the safety stock will be discussed in the next paragraph.

5.4 Safety stock

Safety stock is a protection against uncertainties to prevent from stocking out. Uncertainties result from fluctuations in demand level and lead times. So far both variables have been treated as if they are constant; however in practice they are probabilistic and variable. The demand and lead time fluctuations need to be buffered with safety stocks in order to provide a high service to the customer.

5.4.1 Fluctuations in demand

(29)

Figure 5.2: Safety stock as a buffer against demand variability. Source: http://www.bola.biz/operations/stock/maths.html

There are different methods to set the level of safety stock; for example Silver, Pyke and Peterson (1998) make a rough division in simple minded (see also Schönsleben, 2004), minimizing costs (also Tersine, 1982), customer service (also Vollmann, Berry and Whybark, 1997; Schönsleben, 2004; Hax and Candea, 1984; Tersine, 1982), and aggregate considerations (also Hax and Candea, 1984; Tersine, 1982). For Jan Dekker it is recommendable to base the safety stock on a desired service level, because this is directly related to the purpose of this research and provides a calculateble rule. The preceding authors describe different ways to determine the customer service level; however, for Jan Dekker the aim will be to reduce the probability of stocking out during the replenishment cycle. The safety stock in this cycle service level can be calculated by:

Safety stock = safety factor * standard deviation of the demand during the lead time The safety factor influences service level. The factor can be set equal for each item or be differentiated, for example by using an ABC classification as described in appendix 5. In the case of Jan Dekker three service classes are defined: All products of the supplier Nikko get a reasonably high service level, because the products are of strategic importance to the organization. However, the products do not get the maximum service, because the cost price of these products is very high in general and this would have a too big financial impact. The other products get divided in two groups, based on the holding costs (cost price), so that the total of costs involved are decreased. Setting the safety factor for the items can be chosen intuitively first, and be adapted over time. Figure 5.3 shows the safety factors chosen in this research.

Figure 5.3: Safety factors and service levels of groups of items

Item Safety factor Service level

Nikko 1.9 97%

Low financial impact (30%) 2.4 99% High financial impact (70%) 1.5 93%

For product A the safety factor is set to 1.5, resulting in a safety stock of 18. It is obvious that the order level is influenced by the safety stock and will be increased; The safety stock level of 18 is added to the demand rate during the lead time of 24, resulting in an order level of 42.

(30)

have lead to smoothened or peaked results. Smoothened pattern can result when orders are supplied in subdivided supplies, and peaked results can appear when many backorders are send all at the same moment. However, it is not possible to filter on these abnormalities, therefore demand fluctuation need to be updated regularly with more accurate and recent data.

5.4.2 Fluctuations in supplier lead time

JD also has to deal with supply uncertainties. Fluctuating lead times can result when no ship is available, from slower ships, or customs formalities. Due to the distance and transportation modes, mainly supplies from oversee are sensitive for delays. According to Schönsleben (2004) a safety lead time can be added to the normal lead time to against fluctuations. The lead times at JD are available in the Navision system and are mainly based on the experiences of the purchasers, who know from practice which lead times are plausible. The values cover the whole period from ordering until it is delivered and available for order picking, and include a safety lead time to set off the delivery delays. In this research it is trusted that the Navision data is correct and that JD has sufficiently overcome the fluctuations in lead times.

5.5 Corrections

Constraints can result from either physical or management imposed limitations. If they limit the optimal situation, then adaptations must be made to it. Constraining factors found in literature are the working capital, storage space restrictions and product shelf life (Tersine, 1982; Silver, Pyke and Peterson, 1998). Only the last is possibly restricting for Jan Dekker, because the organization frequently faces product obsolesce due to perishability, causing high costs for the organization. So far the inventory model designed is based on setting decision rules that optimize the costs involved and services level provided, without considering the shelf life.

The technical shelf life on products varies between 6 and 60 months, but the time period in which the item can be sold is shorter. Products are received with approximately 80% remaining shelf life and customers demand 50% remaining. It results in 30% of the technical shelf life is salable; so in case of 6 months, the product can be sold for less then 2 months. In this research the period in which an item can be sold will be named “salable time”.

If the salable time is restrictive, then the inventory level should never exceed the average sales. (Silver, Pyke and Peterson, 1998) However, sales are variable and therefore it is recommendable to even limit the maximum inventory to a minimum level of reasonable sales during the salable time. In literature, no formula was described on this topic. Therefore, a new formula is constructed and applied in this research. The formula is based on the average demand during the salable time, but negatively corrected with a correction factor on the fluctuations on average demand. The correction is comparable to the positive safety factor of fluctuations in calculation of safety stock, but than in inversion. The formula used to calculate the maximum inventory level:

Maximum inventory level =

(31)

correction factor of 1.645. The factor can be chosen depending on the risk that the organization is willing to take. In case of product A, the EOQ plus safety stock is 85, while maximum inventory level allowed results in 199. So for product A an optimal inventory policy can be used. Figure 5.4 illustrates the situation for item A. It has to be noted that it seems that the demand level is totally constant in the figure, but in practice there are variations.

Figure 5.4: Inventory management with no constraints (Item A); Demand during salable time > EOQ + SS

Demand during salable time ≥ EOQ + SS

However, for other products the maximum inventory level can be limiting. This will be demonstrated by the use of three examples with products B, C and D. The limitations for all products are shown in appendix 7.

Limited order quantity

Product B has an EOQ and safety stock of 24 and 8, resulting in a maximum 32 items on stock. However, during the 3.6 months that the product can be sold, the relatively certain demand would be 30. The order quantity can be reduced with 2, so that the risk of obsolesce is limited and the service level can be maintained although total costs will rise.

In this situation the inventory level should never exceed the level of 30, because a higher inventory level would increase the risk of material getting obsolete. Ordering should be up-to the level of 30. This is also illustrated in figure 5.5.

Average demand during salable time

Max. inventory level, reasonably certain demand during salable time

(32)

Figure 5.5: Inventory management with constraints to order quantity (Item B); SS ≤ Demand during salable time ≤ EOQ + SS

Limited service level

Product C has an EOQ of 27 and a safety stock of 12, so the maximum on stock is 39. Based on the demand during the salable time, the maximum inventory level permitted is 10. In this case the maximum inventory is lower than the safety stock required to offer the service level chosen. If it is preferable to limit the risks, then the service level offered must be reduced. The replenishment system will be changed to an order-up-to system, in which the inventory will be filled up to the maximum inventory level. The inventory is replenished up-to this level, because it reduces the average order costs per item against the same risk of obsolesce. Figure 5.6 illustrates this situation.

Figure 5.6: Inventory management with constraint to service level (Item C); 0 ≤ Demand during salable time ≤ SS

Max. inventory level, reasonably certain demand during salable time

Safety stock Time Inventory level Order level Order up to

Max. inventory level, reasonable certain sales during salable time

(33)

The figure does not look like a saw-toothed shape, as in the previous situation. In this case the purchase orders are replenished after each sales order. In the previous situation the stock was consumed more gradually, but neither constant, and replenished after a number of sales orders. Strongly limited service level

Product D’s EOQ and safety stock are 15 and 5, so the maximum stock is 20. Item D has a very low average sales and high fluctuation, because the item is sold but in quit large orders. This causes that negative correction on the fluctuation of sales is larger than the average sales level, and the acceptable inventory level is negative. For products with such a sales pattern it is rather recommendable not to keep them on stock, because the risk depreciation cost would be too high. Nevertheless, there could be other strategic reasons why they should be on stock.

5.6 Overview of the system

In the situation of reasonably variable demand, an EOQ model is a good basis for the inventory management system. Because products have a limited shelf life and the certainty of sales is limited, the maximum inventory levels should be limited to reasonable sales during the salable time of the products. Three exception rules are distinguished from the standard EOQ method to protect against undesired costs for products that risk getting obsolete on stock. Figure 5.7 describes the limitation, and implications of these.

Figure 5.7: Limitations and implications on inventory replenishment system of trade items

Condition

Demand during salable time (Ds)

Order moment Order quantity Safety stock Service level

Standard:

Ds ≥ EOQ + SS

Depends on lead time and average demand

EOQ Depends on standard deviation and service level

Desired level

Exception1:

SS ≤ Ds < EOQ + SS

Depends on lead time and average demand Replenish to level of demand during salable time Depends on standard deviation and service level

Desired level, but against higher costs Exception 2: 0 ≤ Ds < SS At moment of sales Equal to sales order Equal to minimum demand during salable time Reduced Exception 3: Ds < 0 At moment of sales Equal to sales order No No

INGREDIENTS FOR INVENTORY MANAGEMENT TUNING PURCHASE AND SALES AT JAN DEKKER INTERNATIONAL