Working with
Exceptions
Reducing the order processing time at
Karat
Author:
Ale Assies
University:
University of Groningen
Faculty:
Management of Organization
Specialization:
MSc. Business Administration -
Operations and Supply Chains
Student number: 1440462
Date:
Groningen, 6th of July 2006
Supervisor:
Dr. D.P. van Donk
Co-assesor:
Dr. G. A. Welker
Preface
If I explain my specialization, Operation and Supply Chain, to family and friends I always compare it to a puzzle. Every company has its own puzzle. First the pieces lay unstructured on the ground. Each piece can be linked to other pieces by analysing the pieces and looking for the best match. When completing the puzzle there can be some misfits and it is my job to redesign these pieces in such a way that the pieces match better than before.
Karat has offered me a puzzle about order processing. It has pieces like forecasting, order entry, transport and product types. With support of the employees I tried to match the pieces and redesign the puzzle in such a way that it fits better. I would like to thank all the employees who supported me while doing my research and I would like to give some special attention to my supervisors, Mr X and Mr Y, for their support. They have given me the chance to use my theoretical knowledge in a working environment.
I would also like to thank my supervisors at the university, Mr. van Donk and Mrs. Welker, who have helped me with the scientific side of my research. And last but not least, I would like to thank my parents and my friends Eltjo Wannet, Melissande Driesten and Yvonne van der Molen for their sharp remarks to my report.
Summary
For almost a century Karat is a leading company in producing product X. Product X is a product that can be used for many applications. Through the many applications of product X Karat has not only many products but also a diversity of products. When a company, like Karat, produces product X the diversity in products and the various industries can have consequences for the order process. The diversity in products as well as in markets causes the order process to be very complex.
The commercial director and the customer service manager think that the desk managers spend too much time on order processing and that they spend too little time on other tasks, such as prospecting and explore new markets. To find out which factors influence the order processing time a conceptual model is developed. This conceptual model has four factors. These are the operation strategy, the customer/order characteristics, feedback and support and the way the order process is organized. Looking at the management problem and the conceptual model the following main question can be formulated;
Which factors determine the time that is spent on order processing and how can these factors be influenced in such a way that order processing time can be reduced? Which recommendation can be given to get a well-streamlined order process?
After analyzing it is clear that 47,22% of the time of the desk managers is spent on order processing. Furthermore, the four factors can be slit up in several aspects, which have to be changed, to reduce the order processing time. These aspects are; service levels, distribution network, communication to the customer, forecasting, transport, USA products, feedback & support and information systems. With USA products, for instance, the availability of these products is too low and the communication to the desk is moderated. Not all these aspects have the same impact on the order processing time. The sample points out that reducing the amount of orders results in the greatest reduction of the order processing time, reducing the order processing time (without disturbances) has the second priority and reducing the time to solve disturbances has the third priority.
Index
1. ORGANIZATIONAL STRUCTURE... 8
2. PRODUCT PORTFOLIO ... 9
3. RESEARCH METHODOLOGY... 10
-3.1MANAGEMENT PROBLEM... -10
-3.2PROBLEM ANALYSIS AND THE CONCEPTUAL MODEL... -10
-3.3PROBLEM STATEMENT... -12 3.3.1 Subquestions ... 13 -3.4METHODOLOGY... -13 4. OPERATIONS STRATEGY... 16 -4.1PERFORMANCE INDICATORS... -17 4.1.1 Quality ... 17 4.1.2 Speed... 17 4.1.3 Dependability... 17 4.1.4 Flexibility ... 18 4.1.5 Costs... 18 -4.2MARKET STRATEGY... -18 5. ORDER CHARACTERISTICS... 22
-5.1DELIVERY TIME AND THE A/B/C/D CLASSIFICATION... -22
-5.2RUSH ORDERS... -22
-5.3ORDER CHANGING... -23
-5.4DISTRIBUTORS AND AGENTS... -24
6. ORDER PROCESS ... 26
-6.1FORECASTING... -27
-6.2ORDER ENTRY... -28
-6.3MANUAL INPUT... -29
-6.4PHYSICAL DISTRIBUTION... -30
-6.5ORDER ENTRY (USA PRODUCTS)... -30
-6.6RECYCLING... -31
7. FEEDBACK AND SUPPORT... 34
-7.1MANAGING THE PROCESS... -34
-7.2FEEDBACK... -34
-7.3KEY PERFORMANCE INDICATORS... -35
-7.4INFORMATION SYSTEMS... -35
8. CONCLUSIONS OF THE ANALYSIS... 38
-8.1OPERATIONS STRATEGY... -38
-8.2ORDER CHARACTERISTICS... -38
-8.3ORDER PROCESS... -38
-8.4FEEDBACK AND SUPPORT... -39
9. RECOMMENDATIONS ... 39
-9.1DISTRIBUTORS... -40
-9.2DIFFERENT WAYS OF ORDER PROCESSING... -41
9.2.1 Forecasting ... 42
9.2.3 Communication to the customer... 43
-9.3TRANSPORT... -44
-9.4USA PRODUCTS... -45
-9.5FEEDBACK AND SUPPORT... -46
9.5.1 Plan the order process ... 46
9.5.2 Control the order process ... 47
CONCLUSION & DISCUSSION ... 48
REFERENCES ... 50 -APPENDIXES ...I
I.APPLICATIONS OF PRODUCT X...II II.SUB-PROCESSES... III III.STATISTICS ANALYSIS 2005 ... IV
Introduction
Product X is a natural product which can be used for many applications in various industries, such as the food, pharmaceutical and chemical industry. The reason for the many applications of product X is the diversity in products Karat offers. When a company, like Karat, produces product X the diversity in products and the various industries can have consequences for the order process. The diversity in products as well as markets makes the order process very complex.
The general idea at Karat is that this complex situation leads to a situation where the order process takes too much time. The sales department notices that the time for other activities, like sending quotations and prospecting, becomes less. To get more time to do these tasks the order processing time should be reduced. Hence the objective of this research is to find out which factors determine the time that is spent on order processing time and how these factors can be influenced in such a way that the order processing time can be reduced.
The organizational structure and the product portfolio will give an introduction in the world of product X. After describing these aspects the research methodology points out the problem statement, the conceptual model and the methods to find out in which way this problem statement can be solved. The methods which are used in this research are quantitative as well as qualitative. The factors mentioned in the conceptual model are the starting point of the analysis. The analysis describes the main aspects which influence the order processing time. These aspects should be redesigned in such a way that the order processing time will be reduced. The redesign leads to the conclusion of the research and at the end there will be a short discussion about the findings and conceivable further steps to reduce the order processing time.
1. Organizational structure
Confidential
2. Product portfolio
3. Research methodology
The previous chapters have clarified the many possibilities of product X and the reasons why it could be useful for many different companies. As a result of the diversity of customers and products the order process is very complex. This is one of the reasons Karat wants to investigate the order process. In this chapter firstly the problem, formulated by the management team, will be explained. Then the problem analysis, the problem statement and the methodology will be described.
3.1 Management Problem
Karat is one of the biggest suppliers of product X in the world and is always looking for possibilities to expand. The diversity in applications and products makes it important that everyone knows Karat strengths (high quality products and services). The customer service desk should, next to the task of order processing, develop new business and keep the current customers satisfied.
According to the management team the function of customer service desk consists of three elements. These three elements are;
1. order processing
2. prospecting, send quotes, samples, support sales manager 3. maintain relationship with existing customer base
They have the idea that on average each employee should be busy with each of these tasks one third of his time. They have come to this conclusion trough a benchmark in the chemical industry (experiences of other companies). The figures used in this benchmark are based on perceptions of other managers in the chemical industry, which makes this benchmark not really reliable. However, the outcome of this benchmark is the main reason for investigating the order process.
According to this benchmark every desk manager can complete 1300 orders (inclusive task two and three) in a year. If a desk manager only has to keep up with task one he or she can complete 5000 orders in a year. Exceptions are the orders that are being delivered outside the European Union (EU) and recycling orders. Processing this kind of orders takes twice as much time than an order that should be delivered to a company inside the EU. Nevertheless order processing takes, in practice, so much time that there is not enough time left to do the other tasks. On average 6000 orders are being completed in one year. These are the orders from seven and a half Full-timers. Comparing this with the standard it is clear that too few orders are being completed by each desk manager. Furthermore the management team notices that the desk has not enough time to do the other tasks properly. Hence, the management problem is that the desk uses too much time for order processing. The management team wants to know, which factors influence the order processing time and how these factors can be influenced in such a way that the order processing time can be reduced.
The order process in a company is often interwoven with the primary process. The primary process consists of all the processes that add value to a product. At Karat the sales (logistics & customer service desk), production and finance departments are responsible for a specific part of the primary process. Although the whole primary process has to do with processing an order, the focus will be on the order processing activities. Welker (2004) classifies the order process in four different activities; pre-sales activities, order processing activities, order fulfilment activities and the after sales activities. The activities start with forecasting and end with the payment being made by the customer. Activities, which are not classified as order processing activities, will only be discussed if it has influence on the order processing time.
To make a clear definition of order processing time there should be a distinction between the order processing time and the order processing throughput time. The order processing time is the time that is spent on an order actually and the order processing throughput time is the time that all the order process activities have been handled (including waiting time). In this research the order processing time is the time that is spent on the activities, which begins with the order entry and ends when the products are delivered at the customer (and the customers has made the payment). Furthermore goes it only about the time, which is spent through the desk managers.
As mentioned before the management team wants to reduce the time that is spent on order processing. A lot of factors determine the order processing time and therefore it is useful to make a conceptual model. The conceptual model (figure 3) shows the factors that may influence the order processing time. A short analysis of the order process leads to these factors. Through investigating the order process it is possible that it will become clear that some factors have no or little influence on the order processing time. It is also possible that some factors, which are not mentioned in the model, have influence on the order processing time. The model gives only a direction how to investigate the management problem.
Order process
Operations strategy & priorities
Feedback & support Order processing time
Order characteristics
Figure 1; conceptual model
commercial part of the company. How to approach customers? Which are for Karat the most important performance objectives (costs, quality, speed, dependability, flexibility)? The choices that have been made on the operational basis have to match with the decisions which are made on a strategic basis. Therefore is it important to look at the operations strategy of Karat.
The factor described on the right side of the model is the order characteristics. Order characteristics are, for instance, the amount of AOA (automatically order acceptance) orders, rush orders and the reasons why an order should be changed while it is in the process. Also orders from distributors and agents should be handled differently than normal orders.
The factor at the left side of the model is mentioned because the short analysis points out that there are some bottlenecks between some actors of the process. Stock and Lambert (2001) also point out that communication and relations between different actors are important to streamline the order process. The whole order process (with all the sub-processes) will be described and analyzed. Also the disturbances, which occur in the order process, will be described.
The last factor, which probably determines the order processing time, is the way of feedback and support. Harmon (2003) points out that feedback and support (for instance; the information system) influences the order process. In this report only the feedback and support regarding the order process will be analyzed.
All these factors are important to get an overview of the total performance of the order process. In the next section the problem statement and the sub questions will be given. The problem analysis as described in this section is the basis for formulating the problem statement.
3.3 Problem statement
The problem statement is the starting point for this research. A problem statement exists of an objective, a main question and constraints. The management problem is the basis for formulating the objective. According to the management, the problem is that too much time is spent on order processing. This should be reduced to save time to do other tasks, which are related to the customer service desk. Therefore the objective of the research is to know which interventions are necessary to reduce the order processing time.
In the previous section there are factors mentioned in the conceptual model. These factors can possible influence the order processing time and these factors will be used to formulate the main question.
Looking at the objective and the conceptual model the following main question can be formulated; Which factors determine the time that is spent on order processing and how can
these factors be influenced in such a way that order processing time can be reduced? Which recommendations can be given to get a well-streamlined order process?
Constrains can be divided in process and product constraints. The process constraints of the research are the time, namely 22 weeks, the cooperation of the employees and the information which is available in the company, at the university and other institutions.
service desk (desk managers). Production and logistics are not a part of this research and will only be used if it has influence on the activities of the desk. Also finance and other supporting processes are not part of this research. Furthermore, this research goes only about Karat Nederland BV and not about the other parts of Karat NV. For four order types the primary process will be investigated. These orders are an order delivered inside the European Union, one outside the European Union, one recycling order (with two variations) and one order with a product that is being produced by Karat Americas Inc. The last constraint is given by the management and is the fact that the recommendations which are given in this research must be implemented in one year.
3.3.1 Sub-questions
To get a clear answer on the main question there will be some sub-questions. These sub-questions are mainly been developed by the factors giving by the conceptual model. The following sub-questions are given:
1. In which way influences the operations strategy the order processing time? 2. In which way influence the order characteristics the order processing time?
3. In which way influences the design of the order process the order processing time? 4. In which way influences the feedback and support the order processing time?
The answers on these questions are useful to give an answer on the main question. The method how to get answers to these questions is described in the next section.
3.4 Methodology
The previous section gives an outline about what this research wants to find out. In this section the methodology is being given. First there will be a description of what kind of research this is and then the method and sources will be given.
Since the main question is an open question it is an explorative research. Karat does not know which factors determine the order processing time and how these factors influence it. The solutions of this research are only suitable for this particular case and therefore it is a problem solving explorative research. To succeed this successfully it is important to use several sources, to make a chain of evidence and to use key members of the orders process (Yin, 2003). These three key elements are all used in this research to take care of the validity of the research. This research is also reliable, because if another person does this research with the same sources and methods the outcome will be almost the same as described in this report. The outcome will not be exactly the same because some methods, like interviewing, have different outcomes due to the time of when the research took place. Relevance is an aspect that checks if the outcome of this research can be used to make practical changes to reduce the order processing time. The research is practically relevance because of the continue review by the management team and the practical recommendations of this research.
Chapter four till seven analyses the different factors mentioned in the conceptual model. Chapter eight gives the conclusions of the analysis and in chapter nine the recommendations to reduce the order processing time are given.
Task Methods Sources
Introduction Describing the organizational structure
and product portfolio. (chapter 1& 2) - Quality manager - Customer Service Manager - Internet (Robert & Mazonni) Answers on sub-
question 1 Analyzing the current operations strategy with the market requirements. (chapter 4)
- Commercial director & Customer Service Manager - Customer survey report
- Outcome of the statistic analysis (orders 2005) - Literature: Operations Strategy, Slack and Lewis Answers on sub-
questions 2 An outline of how the order characteristics are being used and which disturbances could occur. (chapter 5)
- Desk managers & logistics
- Outcome of the statistic analysis (orders 2005) - Sample orders April 2006
Answers on
sub-questions 3 Modelling the order process with Actor Activity Diagrams. Analyzing the order process by interviewing the desk managers and logistics. (chapter 6)
- Desk managers & logistics
- Literature: Business Process Change, Harmon P. - Sample orders April 2006
- Contracts distributors / agents
- Outcome of the statistic analysis (orders 2005) Answers on sub-
questions 4 Analyzing the feedback and support which are used for order processing. (chapter 7)
- Desk managers & logistics - Key performance indicators 2005
- Literature; Organization development and change, Cummings and Worley
- Information systems Recommendations With support of the diagnosis and
literature some recommendations will be given. (chapter 9)
- Diagnosis - Literature
Table 1; method and sources
The most relevant methods/sources used in this research are interviews, statistics analysis, sample, literature and internal documents. The desk manager, the customer service manager and the people of logistics were interviewed to find relevant information about the factors mentioned in the conceptual model. The statistics analysis about orders processed in 2005 gives an overview about, for instance, the dependability. This was done by dividing the data into the different order types. The literature is used in this research for two reasons. It used as a tool for the analysis (Harmon, Slack) and it is used to support the statements and recommendations in this report. The sample is especially meant for giving extra evidence to the remarks which are given in the interviews. It gives an indication where the main disturbances in the order process are. The sample is done in April 2006 and analyses 76 orders (10 orders per desk manager) of different order types. The sample is split up in 58 normal orders, 14 recycling orders and 4 USA orders. The orders, which are investigated, have already completed the order process. It can be AOA orders as well as orders that are handled manually. In appendix IV a summary of the sample is given and some data are being used in the next chapters. The time to process an order and to solve a disturbance is an estimation of the customer service manager. The sample is translated, with the data of 2005, to how much time it takes to process all the orders (or solving disturbances) of one year.
4. Operations strategy
For almost a century Karat is one of the leading companies in product X. It produces product X for every application, for every branch and for all kind of customers all over the world. Karat can be seen as a supplier with a wide range of products and application services. Karat focuses on all kind of customers and therefore they should know how to organize the order process to meet market requirements. By treating every customer the same and giving all customers excellent service a lot of time will be spent on order processing. It is useful to investigate in which way Karat operates on the market and if this influences the order processing time. In this chapter the operations strategy, the performance indicators and the market strategy will be discussed.
The operations strategy is used in this research because it gives a trade off between the market requirements and the resources. A well-defined operation strategy supports the order process with serving customers on a certain service level. Slack (2002:16) formulates operations strategy as the total pattern of decisions which shape the long-term capabilities of any type of operation and its contribution to overall strategy, through the reconciliation of market requirements with operations resources.
Confidential
Figure 2; outline market segments
Karat formulates in its business plan (2002) that it wants to be strong in three kinds of products, namely the specialty factory products, the commodity factory products and the sourced products. These products are divided in three segments which are outlined in figure four. For all three segments they want to achieve sales targets with a smaller, customer oriented and more effective sales and marketing organization. Logistic goals are factory utilization, costs savings and reducing inventory levels.
4.1 Performance indicators
Slack (2002) indicates five different performance objectives (quality, speed, dependability, flexibility and costs). These objectives will be described with support of the latest customer reports, statistics data and the vision of the management.
4.1.1 Quality
Quality can be formulated as specification quality (Slack, 2002). A product or service needs several dimensions of specification to define its quality. Karat finds the product as well as the service quality important. The dimensions, used in this research, are product quality, the quality of the sales process and the quality of the information/competence. The reason to choose for these dimensions is that Karat has done research (customer survey) to these dimensions.
The customer survey (Newcom Research & Consultancy BV, 2005) shows that the product and the recycling product have respectively an average score of nine and eight, which are extremely high scores. Also the quality of the sales process has a high score, 8.1. When looking at the information and competence of Karat it is show off that these scores are relatively low (respectively 7.3 and 7.4). The survey point out that quality is one of Karat strengths.
4.1.2 Speed
Speed is the time between the beginning of an operations (order) process and its ends (Slack, 2002:44). For A/B products this is 48 hours plus transport time. Because C/D products are made to order the delivery time of these products it can lead up to a maximum of two months (plus transport time). According to the customer survey, customers find the delivery time from Karat sufficient for their process. There are no signals from the market that Karat should improve their delivery time.
Confidential
Figure 3; product evaluation
4.1.3 Dependability
Slack (2002) formulated dependability as keeping up delivery promises. If the actual delivery time is equal to the due delivery time the deviation between these times is zero. Nowadays Karat measures only the requested and the promised date, while they do not measure the date when the product arrives at the customer. Calculating the deviation between the requested and the promised date is suitable for A/B products, but not for the C/D products. This is because for A/B products Karat promised a fixed delivery time and for C/D products not. For C/D products the delivery time is variable and consequently the indicator “promised is requested” is not suitable to measure the dependability. At this moment Karat makes no distinction between the products when measuring the dependability (by Karat called the service level).
(XXX1) (YYY2) World Cust spec. Not cust spec. USA
Amount 3978 892 2131 233 522 250
Service level (all prod.) 82% 74% 69% 100% 88% n.a.
Service level (A/B prod.) 84% 76% 69% n.a. 85% 28%
Table 2; service levels orders 2005
The table above shows that the service level for A/B products, except the recycling orders, stay below the 85 percent. At this moment the dependability of C/D products has not been measured. To measure the dependability of C/D products the actual delivery time should be compared with the promised delivery time. Next to this data the customer survey point outs that ten percent of the respondents do not agree with the statement that the delivery of the products happens on time. Furthermore, the customer survey points out that on time delivery is one of the most relevant aspects. After analyzing this data it can be concluded that dependability is not one of Karat strengths.
4.1.4 Flexibility
Flexibility is a very broad term and has several meanings. Slack (2002) determines four different meanings of flexibility, namely product, mix, volume and delivery flexibility. Through the technical knowledge of Karat they are very flexible in changing products (introducing new products frequently) and the mixing products (easy to change the variety in the product portfolio). Also the customer survey (figure 5) points out that the customers find the product variety a strong point from Karat. Through the technical restriction of the kilns the volume flexibility is limited. The delivery flexibility is also high, because usually they can deliver a rush order on time.
4.1.5 Costs
The last objective that will be discussed is the performance objective “costs”. Because of the high quality and the flexibility that Karat offers, the costs price of the product is relatively high. Often the selling price of traders, like XXXXX and YYYYY, is much lower than the selling price Karat offers. Karat can never compete with these traders, because of the high service of Karat. To get a good product margin the price should include the costs of the high quality products and the application service. Therefore the selling price of Karat is not low enough to compete with the low costs producers. This is in contradiction with what Karat wants to achieve, namely to be competitive in all kind of markets. The customer survey also shows a low score about the prices of Karat’s products. This remark makes a good introduction to the next section were the current situation of Karat is compared to the situation which Karat wants to achieve.
4.2 Market strategy
As said before Karat is as market leader a company that wants to serve all kind of customers. However, different customers have different requirements. For this reason Karat divided their customers in four segments. A short introduction;
High Value
o Loyal customers who order mostly high quality, but expensive products. These customers operate especially in the pharmacy, chemical (specific customers) and water branch and they take care of ? million euro turnover a year.
Base load
o High volume customers with large purchasing power. Price and reliability are the main issues of these customers. These customers operate mainly in the edible oil industry and they take care of ? million turnover a year.
Key competitive
o Customers who fluctuate easily between suppliers. Sometimes they also order with traders. Karat has to prove himself over and over again and price is for these customers very important. Customer operates mostly in gold mines and they take care of ? million turnover a year.
Spot
o Budget buyers where quality is less important. They think commercial and want to have low prices. They take care of ? million turnover a year.
Currently everyone knows this classification, but nobody uses it yet. Karat wants that this year every employee gets familiar to work with this classification. The following table is made together with the management team. They want four different service levels for the four segments. The service levels are meant for all the four product classifications (A-D), except for the dependability.
PERFORMANCE
INDICATORS CURRENT SITUATION TARGETS ALL
SEGMENTS HIGH VALUE BASE LOAD KEY COMPETITIVE SPOT Product
Quality High High High High Low / standard Process
Quality High High High Standard Low support) (no appl.
Speed Standard Standard Standard Low Low
Dependability A/B
Prom. = Req. Low (82 %) High (95%) High (95%) Standard (90%) Low (80 %)
Dependability C/D
Prom. = Actual Not measured 98% 98% 95% 95% Mix flexibility
(quick variation) High High High Standard Low Volume flex.
(changing volume) Standard High High High Standard Delivery flex.
(changing data) High High Standard High Low Costs
(Price) High High High rebate) (with Standard Low
To conclude this chapter it is obvious that the organization is nowadays build for high value and base load customers and not for the key competitive and spot customers. At this moment Karat provides to all the customers the same service. The dependability is divided in A/B and C/D products, because of the differences between these products. Other differences are mainly in the price, dependability and the quality of the products.
5. Order characteristics
Every order that arrives at Karat is treated differently. Some orders came on a regular basis, asking for regular products and some orders are unexpected, asking for irregular products. Although every order is unique the method to process the order is the same. The desk managers treat every order the same, with a little remark that they do a little bit more effort for loyal customers than for customers that order on an irregular basis. In this chapter the delivery time (A/B/C/D classification), the rush orders and the times that an order is being changed will be discussed. Furthermore, the orders from distributors and agents will be analyzed. At the end of this chapter there will be a link of how these order characteristics influence the order processing time.
5.1 Delivery time and the A/B/C/D classification
In chapter two the classification of the products of Karat is explained. This classification is used to determine the delivery time for a specific product. The agreement is that A/B products are delivered in two days plus transport time. These products are already on stock and it is necessary that the factory has two days to put the products ready for transport. The transport time is established with a certain margin. If a customer wants the product earlier than the promised delivery date it is a rush order and then the logistics or traffic manager has to find a solution. Often the transport time can be reduced or alternative transport will be arranged.
The specialty products (C/D) are a different story. The products are not make to stock, but make to order. Delivery within two months is the policy for these products. The exact delivery time for these products can not be given, because it depends if there is sufficient capacity. If the desk manager gets an order for such products the delivery time should be asked to the logistics manager. The logistics manager calls the specific factory and mostly the answer about the delivery time is given two or three days later.
The interviews make clear that not all the desk managers know this classification and if they know they do not use it often. This has as consequences that the customer also does not know this classification. Usually they negotiate the delivery time with the customer in the bidding process. Sometimes they ask for the delivery time at the logistics manager.
5.2 Rush orders
Rush orders occur usually in the Netherlands, France and Germany. The delivery time for customers outside Europe is so extended that it doesn’t matter if the product is a day before or a day after the promised delivery date. In Western Europe the delivery time is much more important. The statistics analysis (table 5) points out that four percent of the orders are rush orders. These are customers that want product from the factories in the Netherlands. From the customers that want product out the United Kingdom only two percent is a rush order.
Parameters Orders EU (XXX) Orders EU (YYY)
Pre-order >= 2 days (A/B products) 96% 98%
Rush order (A,B) <2 days 4% 2%
There desk managers clarify that the three main reasons for a rush order are: 1. The customer places the order too late.
2. The desk manager puts the order too late into the system. 3. Damaged product has to be replaced as soon as possible.
The last reason is mainly the fault of Karat and therefore the only thing that Karat can do is to demand a certain quality from the transport providers. Sometimes the customer has a calamity in their production process or they just order too late. For al lot of customers is product X not the most important product for their production process and for that reason there are negligent in ordering the product X on time.
If they have a calamity the customer doesn’t consider it to be a problem to pay for the extra costs. Sometimes an order becomes a rush order because the desk manager is indolent with putting the order into the system or accepting the order. A strange reason for rush orders is when the order is not arrived at Karat. The customer says that the fax was sent on time, but the fax has never arrived at Karat. A few days later the customer calls with the question; why is the order not yet confirmed? At that moment it becomes a rush order, because the order wasn’t put in the system.
5.3 Order changing
If the order is put in the system there is a chance that the order changes, after it is put in the system. If the order is put in the system the order triggers a lot of actors working at Karat. The logistic manager schedules the order, the factory puts the order ready for transport and the traffic manager arranges the transport. If the customer wants to change the order it will lead to extra work for several people. Therefore the amount of orders, which are changed after putting the order in the system, must be kept to a minimum. In the sample (the table will be explained in the next chapter) the following order characteristics are changed after the order was put into the system; way of payment, customer reference number, delivery date, delivery address and the quantity.
Disturbances Hour / order Amount of orders in sample % of orders in sample Amount of orders 2005 Total time(hours)
Payment 0,25 4 7% 7001 122,52
Missing customer
reference number 0,17 1 2% 7001 23,80
Missing delivery date 0,17 1 2% 7001 23,80
Missing delivery address 0,17 1 2% 7001 23,80
Missing quantity 0,17 1 2% 7001 23,80
Table 5; (part of) summary sample orders EU / world
5.4 Distributors and agents
Besides in the factories, product is also stored at distributors and agents. About eighteen percent of the orders are fulfilled by distributors and fifteen percent are fulfilled by agents. The function of agents and distributors is to support sales with providing extra orders and, in case of a distributor, to have some inventory to handle these orders. Using distributors should lead to fewer small orders and fewer small orders should lead to less order processing time. In this chapter the way of working of the agents and distributors at Karat will be discussed.
The agents have mostly little stock, to suffice the needs of their customers. Usually the transport costs are too high to order little amounts, so therefore they order some extra products. All over the world have agents some inventory. These inventories have no consequences for Karat, because all the risk lay with the agent. The interviews point out that it differs per agent if the order is given on time (and with the right information) to the desk manager. Sometimes the right quantity, price, bank and documents information are missing when the agent is sending an order.
Six years ago Karat began to set up a distribution channel. With the support of distributors they want to serve the customers who order small quantities and save time for the more important customers. Mostly the distributors operate in Europe. The biggest distributors for Karat are Distributor I, Distributor II and Distributor III. These distributors serve customers that order less than tree pallets (about three ton) a time. If these customers however order more than fifteen ton or for more than 25.000 euro in a year, the orders should go to Karat. These agreements come to terms in several contracts with the distributors.
Distributor I Belgium
Agreements supply Æ 10 pallets a time from a factory
Order number 104483 Order number 104304
W20PVS 2 pallets W20PVS 2 pallets
PK251PVS 1 pallets SAE2KPB2 1 pallets
Total 3 PK251PVS 1 pallets
ROW8SPVS 2 pallets
Order number 104527 Total 6
PK13PVS 1 pallets
LQPKVAR 2 pallets
Total 3
Table 6; example Distributor I Belgium
of the contract all the inventory risks lays with Karat. At this moment Karat takes only the advantages of cross selling and they share almost no inventory.
The last conclusion of the interviews is that some parties are agent as well as distributor. In that case the agent operates sometimes as agent and sometimes as distributor. This construction gives some trouble for the desk manager. Often it is not clear if the agent operates as an agent or as a distributor. In this case the agent holds also not much inventory.
After analyzing the order characteristics it became clear that how more disturbances occur in the order process the more time it takes. It not only takes time for the desk manager, but also for the other actors in the process. If Karat wants to reduce their order processing time the amount of rush orders and times that an order has to be changed should be kept to a minimum.
6. Order process
To reduce the order processing time modeling this process is an essential part of the analysis. Modeling the order process clarifies how the sub processes in the order process are organized, who the actors are and what the relations are between these actors. Also it will become clear what the differences are between the four orders types. This chapter gives an overview of the sub processes, the relations and the disturbances between actors.
However, before analyzing the order process, the order processing time Karat used will be discussed. This will be analyzed with the main findings of the sample. The sample (table 8) points out that it differ how much time an order takes. A ‘world’ order takes less than, for instance, a ‘recycling’ order. The table shows that 39,49% of the total time of the desk managers is spent on order processing without disturbances. Solving disturbances takes 7,73% of the time and this leads to 47,22% of the total time from the desk managers, which is used for order processing. The management team wants that only one third of the time should be used for order processing and therefore the time to process an order should be reduced.
Order types Amount of Hour / order Order processing
orders 2005 Time / year
EU order 4870 0,43 2094,10
World order 2131 0,65 1385,15
Recycling order 755 1 755,00
USA order 250 0,86 215,00
Total order processing time 4449,25
Percentage from available time from desk employees Total time (hours)
Total disturbance time normal order 609,09
Total disturbance time USA product 89,38
Total disturbance time recycling 172,03
Total disturbance time 7,73 % 870,49
Total order processing time + 39,49% + 4449,25
Total order processing time (incl. disturbances) 47,22% 5319,74 Total available time employees 100 % 11266,00
Total order processing time (incl. disturbances) - 5319,74
Time left for remaining activities 52,78% 5946,26 Table 7; summary sample
Confidential
Figure 4; ordering process
Figure 6 shows which functions or departments handle the different steps of the order process and it shows the status of an order. When a customer places an order it gets a status in the Karat order information system ORDERLOG. The order start with status one and it is ready when it reached status 99. Before the order entry there are the sub processes forecasting and bidding. These pre-sales activities are very important for the order process, so this is also taken into account. At the end you have the payment done by the customer, but this will not be investigated.
The AAD’s are describing the following sub processes; forecasting, bidding, order entry (normal), manual scheduling, arranging transport, order entry (USA products) and bidding & order entry (recycling). The sub processes are analyzed one by one. What are the relations and what are the disturbances between actors involved in a certain sub process? Only the most important activities of the actors will be discussed. Appendix II gives a more detail description of the activities of the sub processes. Because the unpredictability of the bidding process this sub process will not be discussed.
6.1 Forecasting
The forecast is the basis for the monthly planning. The logistics manager checks the forecast and makes a production schedule. Then the logistic manager discusses this with the factory managers (looking two months in advance). After the planning is discussed the logistic manager freezes it every 15th day of the month. The month April, for instance, will be closed down the fifteenth of
March. Nothing can than be changed on the schedules of production anymore.
To get a good basis for the planning an accurate forecast is very important. No, or an inaccurate forecast can mean that A/B products can not been delivered on time. If the right products are available, it will often lead to an AOA order. An AOA order takes less time than an order scheduled manually and therefore an accurate forecast can lead to a reduction of the order processing time.
6.2 Order entry
The order process starts with an order from a customer that is handled by the desk manager. The main reason for the importance of this sub process is the triggering effect of the order. The order triggers the whole order process.
The order entry starts by putting the order in the order information system and then the order can be accepted automatically (AOA) or the order should be accepted manually. If the order is accepted automatically it takes less time than an order which is accepted manually. The main reasons for no AOA are;
the requested date is too early the order size is too big
there is no stock (that is been forecasted) at the requested date an entity has no AOA code
The first reason comes because customers ordering a product too early (shorter than two days plus transport time). Usually they just order too late, but sometimes it is a calamity in their production process. The third reason comes usually because there is no forecast. If a desk manager puts in the forecast there is a big chance that the product is on stock, if not there is a chance that the product might not be on stock. The last reason is that the product is never produced on stock (mostly a C or D product) and that therefore the order should be always handled manually.
Disturbances Hour / order Amount of orders in sample % of orders in sample Amount of orders 2005 Total time(hours) No errors 0 36 62% 7001 0,00 Payment 0,25 4 7% 7001 122,52 Product availability Rotterdam 0,33 3 5% 7001 115,52 Collections (forwarder customer) 0,33 3 5% 7001 115,52 Wrong product 0,33 2 3% 7001 69,31
Order size too big 0,09 2 3% 7001 18,90
Missing order number 0,17 1 2% 7001 23,80
Order arrives too late 0,17 2 3% 7001 35,71
Freight tables 0,17 1 2% 7001 23,80
Missing delivery date 0,17 1 2% 7001 23,80
Missing delivery address 0,17 1 2% 7001 23,80
Missing quantity 0,17 1 2% 7001 23,80
Wrong info memo 0,09 1 2% 7001 12,60
Total disturbance time 58 100% 609,09 Table 8; summary sample orders EU / world
As mentioned before disturbances takes 7,73% of the total available time of the desk managers. However, disturbances are not only annoying for the desk manager himself, but also for his colleagues at logistics or at the factories. The interviews and the sample make clear that disturbances have to do with product availability, the customer, arranging documents and logistics. This disturbance does not only influence the order entry but also on the manual input and the physical distribution. An accurate planning takes care of the product availability. An accurate planning takes care of the product availability. Disturbances with customers are the following; the customer orders too late, information is missing and incorrect payment. Also documents, transport tariffs and freight tables (transport time) can be a disturbance.
6.3 Manual input
When an order is no AOA the order should be put in manual by the logistics manager. The system puts the order (no AOA) in the list “to be advised” and the logistics manager checks this list a few times a day. The logistic manager looks if there is another possibility to get the order on time at the customer. Regularly the order can get there on time, because there is enough stock or faster transport can be arranged. If there is no stock the logistics manager consults with the desk manager about alternatives. Alternatives are different package, a different product (with the same specification), a new date, different transport, extra costs, et cetera.
the desk manager as well as the logistic manager. There is a chance that, if the order is not accepted fast enough, the products arrive too late at the customer.
6.4 Physical distribution
Arranging transport is done by the traffic manager and the logistics employee (only sea transport). This transport (plus documents) will be arranged if the order is accepted. Karat uses external parties to arrange internal as well as external transport. Two transport providers (Provider I and Provider II) arrange the transport for Western Europe (17 countries) and for the rest of the world Karat uses several transport providers (road, air and sea transport.). Every day logistics have contact with the transport providers as well as with the factories. Two times a day some lists with orders that should be handled are forwarded by e-mail to the transport providers. These times (around 8:30 a.m. and 14:00 p.m.) are extremely important, because it can influence if an order will be on time or not. The rest of the day is used for trouble shooting and arranging documents. The interviews and the sample make clear that there are three main disturbances in this sub process. These are the freight tables, documents and the quality of the transport providers.
Freight tables are calculated transport times from the factory to a certain area. Freight tables are mostly a problem in France and Eastern Europe. The freight tables are too wide and therefore the order will not automatically be accepted. The freight tables in Eastern Europe are too wide because the tables are calculated for the situation before the countries belong to the European Union. In that old situation the trucks has huge waiting times at the border and nowadays this is not the case anymore. In France the product form London is a problem. This product should go with the ferry (which leaves two times a week) and therefore there is some slack build in the freight tables. The problem is that the customer finds the transport time too long and wants the products earlier. At that moment the desk manager should arrange this with the logistics manager. Usually a shorter transport time is possible, but it takes extra order processing time to arrange this. When the transport time will be shorter calculated the order will be accepted automatically, which takes less order processing time. For the countries in Eastern Europe the freight tables can be changed, but the situation in France is more complex. The ferry and the fact that product from London are hazardous goods makes it difficult to change the freight tables.
Problems with documents usually have to do with documents that have to come back from external parties (customs, shipping company, etc). This can give problems for the customer to get his products. The last disturbance, the quality of the transport providers, is a serious problem. The customers in West Europe complain regularly about product damage, loss of documents and the delivery date that differs from the date that is promised by the desk. Especially the product from London gives problems. The provider who takes away products form London communicates badly to the customer as well as Karat. If, for instance, the provider is too early at the customer they don’t communicate this.
6.5 Order entry (USA products)
orders every time a “full container load” to get some stock in the factories. In this case the order entry is the same as products that are produced in the Netherlands.
It is different when a desk manager gets an order with a quantity that contains a full container load. Now the desk manager orders the container directly from America. This time the products will not be stored in the factories in the Netherlands, but directly be delivered to the customer. The activities that should be done by the desk manager take two times more time than a normal order. This has to do with the different information systems, XXX and ZZZ (explanation in section 7.4). The desk manager should put the order in both systems. In ZZZ the desk manager puts almost the same information as in XXX, but some things differ. Take, for instance the incoterm, the quantity (lbs instead of kg), the currency (USD instead of EUR), customer records, et cetera. The double import of the order takes much time, especially because some aspects (for instance the quantity) must convert to the situation in America. The chance to make mistakes is also become bigger, due to the double import of order information.
Disturbances Hour / order Amount of orders in sample % of orders in sample Amount of orders 2005 Total time(hours)
No errors 0 0 0 250 0
Documents USA orders 0,42 3 75 250 78,75
Product availability USA 0,17 1 25 250 10,63
Total disturbance time 4 100 89,38
Table 9; summary sample USA orders
It is worrisome that from all the orders of 2005 that go directly from America to the customer only 28 percent (table 3) of the A/B products are on time. The reasons for this shocking service level are the technical restrictions of the American factories. Therefore, the products are often not available (table 10). The interviews point out a chance that, in spite of a forecast and setting the order on time in the system, the order comes too late at the customer. Sometimes they promise to deliver the product to a customer from Karat Nederland BV, but at last they prefer their own customer. Also the bad communication about the delivery time is very annoying for the desk manager. Not one desk manager is satisfied with the cooperation with Karat Americas Inc.
6.6 Recycling
Recycling orders can be separated in customers that want their own reactivated product back (customer specific product) and customers that do not care which product they get back (not customer specific product). The customer specific recycling is difficult to schedule since the product must not be mixed with other products. Orders from customers who do not want their product back are easier to schedule, because it can be mixed with other products.
Confidential
Figure 5; example recycling order
In contrast to the transport of a normal order, Karat arranged self the transport for recycling. Karat has a number of trucks to get the product from the customer and bring the product (which is reactivated) back to the customer. To get an idea of this complex situation the above example (figure 7) is given. If a customer wants to recycle 150 m3 product (150 m3 is the maximum amount
that a customer can recycle), six tank carriages (25 m3 a tank carriage) are necessary to get the
wet product from the customer and three carriages (50 m3 a tank carriage) to get the dry product X
back to the customer. The logistics manager arranges the tank carriages and the recycling manager (who takes care of the recycling process in the factory) arranges the capacity of the silo’s / recycling kilns. When the recycling manager schedules the recycling order the tank carriages planning is also taking into account. The desk manager put just like a normal order the recycling order into the system. The only differences are that the desk manager should contact the recycling and logistics manager before the (load and unload) dates are put into the system.
Although a recycling order is more complex than a normal order, every recycling (customer specific) order from 2005 (total 233 orders) is on time. A service level (promised = requested) of 100 percent is excellent. The not customer specific order goes a little different. The not customer specific product is called BAS and is transported by truck combinations. This is a combination of a tank carriage and a container. In the tank carriage new product is delivered and the old product goes back in the container. This way one truck can deliver new product and take the old product back. Sometimes a customer delivers only the old product or they buy only BAS. Karat has three truck combinations and together with arranging licenses3 these are the bottlenecks for delivery on
time. Therefore the management finds a service level of 85 percent acceptable.
Disturbances Hour / order Amount of orders in sample % of orders in sample Amount of orders 2005 Total time(hours)
No errors 0 5 36% 0
Capacity truck carriages 0,42 7 50% 755 158,55
Missing order number 0,17 1 7% 755 8,98
Delay at customer 0,085 1 7% 755 4,49
Total 14 100% 172,02
Table 10; summary sample recycle orders
The interviews and the sample show that logistics as well as desk managers find it sometimes difficult to schedule the recycling orders (especially BAS). Half of the orders used in the sample have scheduling difficulties. Reasons for these scheduling difficulties are the limited capacity and the short delivery times that are arranged in the contracts. In that last case it is very difficult to organize that the product is picked up and deliver in just a few days. Although it is often possible to find a solution, it takes a lot of extra order processing time.
3Licenses are meant for getting the old carbon (cross-border detritus) across the boarder. For every order outside the Netherlands
7. Feedback and support
Managing and improving the order process is a complex process. This becomes even more complex when the markets are diverse and difficult to compare. Karat operates all over the world in several branches with different characteristics and a customer from France is very different than a customer from Slovakia. For these reasons the feedback and support should be designed is such a way that the employees handle the order as efficient as possible. This chapter describes the way the desk is managed, the process is measured, how the feedback is organized and how the information systems support the order process.
7.1 Managing the process
The customer service manager is responsible for the order process. When managing the order process the managers should take the technical restrictions into account. The management takes care of the internal training, measuring the process and giving feedback by means of Key Performance Indicators (KPI).
The employees work independently from each other and they all have own responsibilities. Although everyone likes this independence, there is confusion about the order processing policy Karat uses. At this moment all the employees do whatever they think they should do, but there is no manual for order processing. The only thing that is available is the ORDERLOG (order processing information system) manual, but that only takes care of the technical aspects of the information system. Because the lacking of certain rules the internal training of new employees differs. It depends on the previous employee’s function how a new employee is trained. Some new employees find the internal training not sufficient to do their work properly. A positive point is that the employees are satisfied with the way that the management handles their remarks.
7.2 Feedback
One way to continue to improve the process is giving feedback. Cummings and Worley (2001:132) describe feedback as meetings that provide a forum for discussing data, drawing relevant conclusions and devising preliminary action plans. Feedback can have negative and positive consequences, depending on how the feedback is organized.
7.3 Key Performance Indicators
KPI’s are used by the management to formulate their goals and to measure the order process. The indicators are only used for the areas in Europe and it has to do with the service level (promised = requested) and the amount of orders that are accepted automatically. Table 12 gives a summary of the actual performance and the targets in 2005.
As mentioned before AOA takes less time than a manual scheduled order. In the KPI’s the amount of AOA for every area is described and this is also divided in products that are produced in one of the three factories (London, Eindhoven and Rotterdam). The reasons why orders are handled manually are also mentioned.
2005
Actual service level (%) Target service Level (%) Actual AOA (%) Target AOA (%)
January 80 80 52 60 February 81 80 61 60 March 84 80 56 60 April 82 84 56 67 May 80 84 58 67 June 80 84 79 67 July 83 88 70 74 August 80 88 61 74 September 85 88 60 74 October 88 90 61 80 November 83 90 61 80 December 85 90 63 80
Table 11; key performance indicators 2005
Another performance indicator is the service level Karat uses. This is a comparison of the requested and the promised date. If all the orders have the same requested as promised date the service level will be 100 percent. As mentioned in chapter four the dependability is too low. One reason might be the classification of the products. The service levels takes all the products into account and this service level (promised = requested) can only used for A/B products and not for C/D products. For C/D products the service level, promised = actual delivery, is more useful.
7.4 Information systems
Nowadays information technology is indispensable in every company. The information system that supports the order processing function is designed in such a way that it provides a communication network that links the customer and the supplier (Stock & Lambert, 151). Karat uses several information systems to support this function. In this section a short overview of the information systems (table 13) Karat uses will be given. Furthermore the comments of the employees who work with the systems will be given.
Another remark is that information is put on several places and therefore it is sometimes hard to find the right information. Customer information, for instance, is saved on three different places. Quick finding of information is one of the issues, which can reduce the order processing time. Especially for the desk managers, who are providing information to the customer, quick and correct information is critical for an efficient order process.
Information system Function
ORDERLOG This order processing program is used at Karat since 1996. It’s a MS-DOS environment, where the user can’t use several fences at one time. To trace the progress of the order it gets every time a different status. This system has several functions. An employee can, for instance, read the inventory levels. All the actors of the primary process use this system.
Pro100 Pro5 is a graphical, Windows like, environment and is coupled to the data, which is used in ORDERLOG. This system is three years old and it contains information about transport, customers and bulk planning. The desk can put in different customer data, like quotations, forecast, complaints, contracts, et cetera. It can be used to look for operational and management data.
Outlook The mail program Karat uses is not coupled to the other systems and the correspondences with the customer (orders, offers, etc) are saved in public folders.
Intranet Here are some supporting tools, like product specifications, organizational diagrams, brochures, et cetera. Also the quality system is put on the intranet.
MS Query An database with all the information that is put in ORDERLOG or Pro100
Table 12; information systems
ORDERLOG has three different sub systems, namely XXX, YYY and ZZZ. These systems look the same, but there are some small differences. One difference is that the desk manager should put on order for a product from the United Kingdom in the YYY system, an order for a product from the Netherlands/ Italy in the XXX system and an order for a product from America in the ZZZ system and the XXX system. These differences are designed, because the three different countries (USA, United Kingdom, the Netherlands) has there own parameters for weight, currency, et cetera. The results of this differences is that the double input (in ZZZ and XXX) on one order enlarges the chance for human errors. Another problem of ORDERLOG is that some of the functions can not be used by several people at one time. For instance, if the traffic manager wants to make a list for the transport providers, all the desk managers can not put an order into the system.
Another point, mentioned in the interviews, is the large amount of errors in Pro100. These errors happen frequently and it results in loss of time for the desk manager. A reason for these errors is the fact that a lot of data (in two systems, ORDERLOG and Pro100) is coupled with each other. This results regularly in a stagnation of the system.
The last point is the inconsistency with saving documents (like orders, quotations) in the public folders. The agreements that are made a year ago about saving orders and offers are sometimes complied very strictly and sometimes not. The problem of incorrect documentation is that it is more difficult to find the right information. This could results in more order processing time.
8. Conclusions of the analysis
After analyzing the order process some main conclusions can be written. These conclusions are the starting point for the recommendations of this research. The conclusion of the analysis is that all four factors of the conceptual model have influence on the order processing time. The analysis makes clear that the four factors can be slit up in several aspects, which has to be changed, to reduce the order processing time. The four factors of the conceptual model will be discussed in the next sections.
FACTORS CONCEPTUAL MODEL NEW FACTORS
Operations strategy Service levels
Order characteristics Distributors, Communication to the customer
Order process Forecasting, Transport, USA products
Feedback and support Feedback and support, information systems
Table 13; factors
8.1 Operations strategy
The conclusion, which can be made through analyzing the operations strategy, is that Karat provides the same service for all their customers. This has consequences for Karat’s overall performance. Karat has already divided their customers in four different segments and now it is important to make four different service levels. These service levels have to be communicated to the customer. When using the service levels the policy of how to approach customers is clearer and that can result to some reduction of the order processing time.
8.2 Order characteristics
Chapter seven points out that the function of the distributors of Karat is mainly cross selling. They do not have much inventory, which is often the function of a distributor, contracts are not followed upon and the agreements in the contracts vary with each distributor. The distributors take up as much time as a normal customer and that can not have been Karat’s idea when it was setting up a distribution network. The order processing time can be reduced through changing the distribution network
The policy of how Karat deals with order characteristics (like rush orders, order from distributors et cetera), should be communicating to customer. At the moment the customer does not know the differences between A/B and C/D products. Also the policy about rush orders is not clear.
8.3 Order process
As mentioned in chapter six forecast and product availability are closely related. Forecasting is the basis for the production planning. According to the desk managers forecasting is, especially in certain areas, extremely difficult. To get enough inventories of the MTS orders and to meet the customer’s specification the forecast should be correct. If the A/B products are not on stock Karat does not meet the customer expectations and they should look for alternatives. To have the right products available the order processing time can be reduced.
