Winning More Projects

(1)

Winning More Projects

Improvement of the Quotation Process at Draka E&I Europe

Master thesis, Msc Business Administration, specialization Operations & Supply Chains University of Groningen, Faculty of Economics & Business

(2)

2

Abstract

(3)

3

1. Introduction

In order to increase company sales, the management of the division Energy & Infrastructure Europe of cable manufacturer Draka wants to improve its position in the international, industrial project market. At the moment, the division is not successful in this objective. Not many industry project orders are won. The division management indicates that one of the reasons for this is the lack of responsiveness to customer order enquiries. It often happens that Draka does not respond to project enquiries with a quote because the quote cannot be prepared in time. In other cases Draka does make a quote, but the chances of winning the order reduce when the customer deadline is exceeded. This research therefore will focus on the responsiveness of Draka E&I Europe to project enquiries. In this section first relevant information about the company will be provided. Subsequently the process that is studied will be introduced briefly, followed by the formulation of the management problem and the problem statement.

1.1 The organization

The subject of this thesis is the European division of the Energy & Infrastructure (E&I) group of Draka Holding NV. Draka is a cable manufacturer that is operating world-wide. Draka has divided its product market in three groups: Energy & Infrastructure, Industry & Specialty and Communications. These groups again are subdivided in different divisions, as can be seen in Figure 1.

Figure 1: Organization chart

E&I Europe develops, manufactures and markets low-voltage and medium-voltage cables. The product range extends from installation cable and PVC flexibles to power-, instrumentation- and control cables. These cables are delivered to different customers. Around 80 percent of the sales volume is directed to wholesalers, who take care of the distribution of the cables to the end-user. Next to the wholesale market, Draka wants to be active in the industrial market. A difference between the wholesale market and the industrial market is that for the industrial projects often

Draka Holding NV

Energy & Infrastructure Industry & Specialty Communications

Europe

Asia-Pacific

Automotive & Aviation

Cableteq USA

Elevator Products

Industrial

Telecom Solutions

Multimedia & Specials

Americas

(5)

5 customer-specific cables are demanded, while wholesalers mainly demand standard products. When customized products are demanded, these products need to be designed at one of the product design departments of the division.

1.2 The quotation process

The process of responding to customer enquiries is called the quotation process. Bramham et al. (2004) define the quotation process as the process of converting a description of customer needs into organizational capabilities. This means that the quotation process is initiated by a customer who submits its wishes and demands to Draka. The output of the quotation process is a quote: a document that describes the commitment by a business to the customer in terms of product specification, price and delivery. Customers usually request quotes at multiple companies, so they can choose the most attractive offer. When customized products are demanded, the quotation process includes the designing and price calculation of the demanded cables. When standard products are demanded, the product specifications and prices are already known, so these steps are not included in the process. A characteristic of the international projects Draka wants to participate in is that the customers hardly ever demand cables from the standard assortment of Draka.

The quotation process of Draka E&I Europe can involve departments in multiple countries. This is a consequence of the focused factory philosophy (Skinner, 1974) that is introduced several years ago. The division has separated the different sales offices and specialized factories throughout Europe. The different factories are responsible for the production of different parts of the product range. The consequence is that the holding partners do not produce all products for their market themselves anymore, but that they have to be supplied partly by holding partners in another country. Intercompany trade, selling cables to and buying them from other Draka locations, became more and more important. By dividing the product range over different factories cost advantages were achieved: as a consequence of the fact that the factories became responsible for the production of more similar products, setup times were reduced and production efficiency increased. While production was spread over Europe, the sales organizations in the different countries still approach their market locally. Because of this situation, it can happen that customers of the UK sales organization demand cables that are produced in the Netherlands. In this situation, the quote has to be prepared together by the English and Dutch organization. The three countries that are mostly involved in the quotation processes for projects are UK, Spain and the Netherlands.

(6)

6 Project Enquiry received Quotation made Period in working days Urea plant 24-9-2008 28-10-2008 28 days

MEG plant 12-3-2009 20-4-2009 28 days Oil plant 10-2-2009 13-3-2009 24 days Power plant 23-2-2009 7-4-2009 32 days Steam generator 26-3-2008 6-5-2008 30 days

Table 1: Sample of previous project quotation processes

Although these cases are just a sample of a larger total of enquiries, they confirm that Draka sometimes has problems answering to customer enquiries within fifteen working days. Furthermore, there are many project enquiries Draka does not even respond to, because it would take too long to prepare the quote. As this situation is considered undesirable by the division management, the international quotation process of Draka is the subject of this study.

1.3 Problem statement

The division management indicates that currently not many project orders are won as Draka is unable to deliver the quote to the customer in time. The management problem can be summarized as follows:

Draka is not responsive enough to project enquiries, which causes that Draka cannot improve its position in the project market

This management problem leads to the following research question:

Which factors influence the quotation process of Draka E&I Europe, and how can these factors be changed so the responsiveness to customer enquiries increases?

This research will focus on the customer enquiries that are processed at the Dutch product design department. The vast majority of these enquiries are initiated by the sales organizations in the UK or Spain. Because the Netherlands is the most important production location for the international projects, it is appropriate to limit the research just to these enquiries. Because the research is performed in the Netherlands it is also the most practical to confine the research in this way.

(7)

7

2. Research methods

The upcoming section will discuss the different ways of data collection that are used for this research.

The research method is a literature study combined with a case study. To be able to answer the main research question, first relevant literature was studied to find out which factors can influence quotation processes. Furthermore, literature related to managing product variety and planning and scheduling issues was studied as these issues are related to the main research problem. Several online scientific databases were used as source for the articles referred to in this paper.

Next to the literature study a case study at Draka was performed. A number of sources were used to collect information about Draka and the quotation process. Firstly, different forms of secondary data were analyzed. These data included annual reports, process manuals, internal and external presentations and data from the ERP system Baan. In addition to the secondary data, several interviews were executed with different persons. In order to get a better understanding of the Draka organization in the Netherlands interviews with the managing director, the division president and the divisional marketing manager were conducted. To get more insight in the activities specifically related to the quotation process, interviews were held with the intercompany manager, the project manager and the business manager of the Dutch organization. The product design manager and two employees of the product design department in the Netherlands were interviewed about the jobs and tasks at that department, and about the method of planning of these jobs and tasks. Furthermore an observation at the product design department was executed in order to get an insight in the practice at the product design department.

As the studied process involves more organizations than just the Dutch organization, also interviews were conducted with involved employees of holding partners abroad. Phone interviews were conducted with the project manager and the sales manager from the UK, and the project manager from Spain. Finally, a number of meetings from the industry projects core team, that consists of business managers, project managers and sales managers from Spain, UK and the Netherlands) were attended.

Validity of the data

(8)

8

3. Theoretical framework

In this section an overview will be presented of the literature that is relevant for Draka’s problem. The goal is to provide an overview of the factors that can affect the responsiveness of Draka’s quotation process for customer-specific products. The sub questions will be derived from this literature study, and they will be answered in Chapter 4. Together they will lead to an answer to the main research question.

The problem that Draka faces is that the throughput time of the quotation process often is too long. To address this issue, first relevant literature regarding the management of quotation processes will be discussed. Subsequently the influence of variety on quotation processes will be analyzed. Finally will be discussed what factors influence the responsiveness of an organization in general.

3.1 Managing quotation processes

Bramham et al. (2004) analyzed the quotation processes of two businesses in order to come to a better understanding of quotation processes in general. A similarity of the two businesses is that both use a product configuration system (Forza and Salvador, 2002), in contrast to Draka. A product configuration system can be defined as an IT-system capable of combining well defined building blocks governed by rules and constraints into a product. This system can be used by the different departments involved in the quotation process, to make sure that the customer’s demands match with the production capabilities. At Draka, only at the product design department is examined whether this match exists. There is no such system available for the sales organizations to guide the quotation process and judge the feasibility of customer’s demands.

The authors conceptually analyze the two quotation processes and identify four key process stages and four key decision areas of quotation processes. Their conceptual model of the quotation process is shown in Figure 1. The four central elements that are connected by arrows represent the four process stages; the numbered elements are the different related decision centers.

Figure 2: Bramham et al. (2004) quotation process framework

(9)

9 Decision center Aims of decision center

I) Customization request initiation and information gathering on customer needs

• Collect information in dialogue with the customer on their requirements

II) Classification of requests _{• Route enquiries to the relevant experts in the} company

• Understand the scale of the modifications to meet customer requirements

• Recognize the closest match product that might be “cannibalized” to meet customer needs or initiate new product development to meet customer needs III) Resource control _• Assign resources to the consideration of customer

requests

IV) Identification of information for reuse _• Assess what information is likely to be useful in the future for further customer orders or quotations • Analyze the feedback on the success of quotations

and accuracy of estimates associated with customization requests

Table 2: Overview of decision centers

The authors indicate that in certain business environments some of the decision centers may be less significant, but the classification decision center is identified as one of the most important decision centers for effective and efficient quotation processing.

As the performance of the quotation process depends on the functioning of the four decision centers, process improvement should be achieved by improving the functioning of these decision centers. Therefore, the following sub question is formulated:

Is it possible to improve the performance of the four decision centers of Draka’s quotation process?

To find out which aspects of the decision centers can be improved, first the factors that can influence the performance must be identified. Therefore, in the upcoming section the literature about the factors that are relevant to Draka’s quotation process are reviewed.

3.2 Improving responsiveness

The problem that Draka has with responding to customer enquiries can be characterized more in general as a responsiveness problem. Reichhart and Holweg (2007) come to the following definition of responsiveness:

The responsiveness of a manufacturing or supply chain system is defined by the speed with which the system can adjust its output within the available range of the four external flexibility types: product, mix, volume and delivery, in response to an external stimulus, e.g. a customer order.

(10)

10 Responsiveness thus is determined by the level of flexibility, where flexibility can be any of the four mentioned variants. Slack (2004) defined the four types of flexibility as follows:

Flexibility type Definition

Product The operations ability to introduce new or modified products and services

Mix The operations ability produce a wide range or mix of products and services

Volume The operations ability to change its level of output or activity to produce different quantities or volumes of products and services over time Delivery The operations ability to change the timing of

the delivery of its services or products

Table 3: Overview of flexibility types

These definitions imply a process that has a product or a service as output. The output of the process that is studied in this paper however is not yet the final product; it is the quote for the customer. Keeping this in mind, the flexibility types can be translated to the specific situation of a quotation process. Product flexibility is the ability to deliver new quotes to customers. These are quotes for products that Draka never made a quotation for. Mix flexibility is related to the ability of producing a wide range of different quotes, i.e. quotes with a wide range of different products. Volume flexibility has to do with the ability to change the number of quotes that are produced by Draka. Finally, delivery flexibility is about the ability to reduce or extend the time that is needed to fulfil the quotation process.

The lack of responsiveness at Draka is related to the level of product flexibility, as the problem with responding to enquiries occurs when new products or modifications to existing products have to be offered to the customer. Also volume and delivery flexibility play a role. Volume flexibility is low because at the moment Draka is not able to heavily increase the level of quotes that is created. Delivery flexibility is low, because Draka is not able to increase the speed by which the quotations are produced. As Draka performs insufficiently on multiple forms of flexibility, its organization problem can be defined as a responsiveness problem. Therefore it is interesting to find out which factors are recognized in the literature that affect the level of responsiveness of an organization.

Reichhart and Holweg reviewed the existing literature on responsiveness. As a result, they formulated eight determinants for responsiveness, which are divided in operational factors and supply chain integration factors:

Operational factors Supply chain integration Demand anticipation Information integration

Manufacturing flexibility Coordination and resource sharing Inventory Organisational integration

Product architecture/postponement Spatial integration and logistics

Table 4: Overview of determinants of responsiveness

(11)

11 determinants that Reichhart and Holweg found may still influence the level of responsiveness of Draka. The operational factors refer to the individual actors in the supply chain, which means that they can also be applied to an individual organization in the Draka division. However, the operational factors apply in particular to real production situations. The supply chain integration factors may be relevant because the process involves different organizations. These organizations, like partners in a supply chain, equally require information exchange, coordination and integration in order to complete the quotation process. Therefore the next sub question is formulated:

To what extent do the determinants of responsiveness affect the responsiveness of Draka’s quotation process, and how can the determinants be used to improve the responsiveness?

3.3 Managing product variety

Draka’s lack of responsiveness to enquiries is especially related to the demand of non-standard products by customers. When customers demand standard products, the information that is needed to make a quote is already in the information system. In these cases Draka has no problems to reply to the customer in time. However, customers demand a wide variety of products. Besides the generic model for quotation processes, Bramham et al. (2004) also discuss the management of quotation processes with high product variety in their article. Two mechanisms of dealing with variety in the quotation process are identified: absorbing variety and mitigation of variety. Table 5 provides an explanation of the two methods.

Mechanism Purpose Methods

Absorbing variety Deal with variety in demand in the best possible way

• Allocate resources to customer requests in order of importance

• Meet customer’s needs with a product specification of closest appropriate fit • Update information systems with efficient

use of resources

• Monitor changing delivery capabilities according to the product variety loading Mitigation of

variety

Try to reduce the level of product variety in the quotation process

• Minimize the proliferation of variety • Filter requests by customers for

non-standard products

• Rejection of customer requests for infeasible or “inappropriate” products • Monitor customer requests in terms of

types of enquiries and orders

Table 5: Overview of methods to manage product variety

The described methods aim at improving quotation processes that have to deal with high variety in demand. This is relevant to Draka’s situation, as customers of Draka demand a wide variety of often customer-specific cables. Applying these methods may therefore improve the functioning of the decision centers in Draka’s quotation process. This leads to the following sub question:

(12)

12

3.4 Conclusion

Draka’s quotation process should become more responsive to customer enquiries. The model of Bramham et al. (2004) showed that a quotation process is composed of four process stages: Understanding of customer needs, consideration of the customer request, generation of the quote and update of the information system. The functioning of these process stages is influenced by four decision centers: initiation of the customization request, classification of the customer request, resource control and identification of information for reuse. Therefore it is vital to design these decision centers as efficient as possible when you want to improve the performance of the different process stages.

Several factors were found that can influence the functioning of the decision centers of the quotation process. Two methods of managing product variety in the quotation process were identified: absorption of variety and mitigation of variety. Next to these two methods, certain determinants for responsiveness in general were found that may affect the performance of the quotation process in this specific case. In the remainder of this research the impact of any of these factors on the performance of the different decision centers of Draka’s quotation process will be investigated.

(13)

13

4. Results

In this chapter, the presumed relations that are described in the theoretical framework will be tested. First a schematic view of the current quotation process will be provided. Subsequently the functioning of the four decision centers will be elaborated, and the impact of the different influencing factors will be discussed.

4.1 The quotation process

Figure 4 provides a schematic representation of the current quotation process. The blocks on the right represent the different process steps. The blocks on the left show which person or department is involved in these steps. The figure is applicable to the enquiries that are received by the sales organization in the UK or Spain, and that contain products that are produced in the Dutch factory.

L o c a l s a le s ( U K /S P ) P ro je c t m a n a g e r N L D e s ig n d e p a rt m e n t N L P ro je c t m a n a g e r N L L o c a l s a le s (U K /S P )

(14)

14 The above figure provides a general overview of the quotation process. In reality the course of the process may deviate from this representation. The different steps of the process will be discussed more in detail in the upcoming section, based on the four decision centers of Bramham et al. (2004).

4.2 DC 1: Customization request initiation and information gathering on customer

needs

The quotation process is initiated by an enquiry that a potential customer sends to a local sales organization. The enquiries that this research focuses on are received by the sales organizations in the UK or Spain. The enquiry is a set of documents that contain information about the cable types and the specifications that the customer wants to order. The first review of the enquiry is performed by the sales organization that receives the enquiry. The salesperson checks what cable type is requested, and when it turns out that the customer requests cables that are produced in the Netherlands, the enquiry is forwarded to the Dutch project manager. At this stage of the process, the sales organization does not communicate with the customer regarding his detailed requirements. The communication about the exact customer demands starts when the product design department reviews the enquiry, the third review in the quotation process. During the technical review at the design department questions or uncertainties regarding the customer’s demands may arise. When the product designer requires more information about the customer’s demands, the sales organization is informed who in turn contact the customer. The information exchange between the customer and the product designers thus does not happen directly, but always via the sales organization. Figure 5 provides a representation of the communication process.

Figure 5: Representation of the communication process

The numbers in the figure indicate the sequence of the information flows. The situation modeled in the figure assumes that the customer is consulted only once regarding his exact needs after submitting the enquiry. In practice the customer often needs to be consulted multiple times, which means that steps 3, 4, 5 and 6 are repeated.

(15)

15 demands can take several weeks. As the total quotation process may take no longer than three weeks according to the KPI, this extensive communication is an important factor that affects Draka’s responsiveness.

4.3 DC 2: Classification of requests

The classification stage is acknowledged as one of the most important stages in the quotation process. During this stage the feasibility of the enquiry is assessed by internal experts, and eventually the decision is made to accept or to reject the enquiry.

The first assessment of the feasibility is made by the sales organization that receives the enquiry. The sales person determines which cable type is requested, and makes the first judgment whether the cables can be produced in one of Draka’s factories. The next important decision is the decision of the sales organization to forward the enquiry to the right production organization. In order to be able to make this decision correctly, the salespersons should be well informed of the production capabilities of the different factories throughout Europe. This decision therefore requires a high degree of coordination of the different sales and production organizations.

When the project manager in the Netherlands receives the enquiry, he further reviews the enquiry on a number of criteria, and then decides whether the enquiry should be forwarded to the design department. The project manager bases this decision on two factors: the commercial viability of the enquiry and the work pressure at the product design department. The project manager then decides whether the Dutch design department is going to process the enquiry or not. In around 50% of the cases the project manager decides to refuse the enquiry. The project manager states:

“I have thirty years of experience in making quotes for projects. Therefore I can decide whether it is useful to quote for a certain project. In about fifty percent of the enquiries I receive I decide not to make the quote. UK and Spain focus on acquiring as much enquiries as possible; there is no point in making offers for all these projects.”

The above citation of the Dutch project manager indicates that the sales organizations in the UK and Spain and the Dutch production organization have a different view on the project market and the best way to approach it. This suggests that there is a lack of organizational integration within the European E&I division. One of the main characteristics of organizational integration is the alignment of organizational goals (Chen et al. 2009). It appears that the sales organizations in the UK and Spain want to focus on growth on the project market. In the Dutch production organization, the focus is primarily on the wholesale market. The consequence is that the enquiries are classified differently by the different organizations. Many enquiries that are assessed as commercially viable by the sales organizations are rejected by the project manager in the Netherlands.

(16)

16 conversations between product design and sales about the exact demands of the customer show the need for extra communication after the initial submission of the enquiry. This indicates that the information gathering about customer needs and the classification of the enquiry are simultaneous processes in Draka’s quotation process, while the model suggests that the different decision centers take place consecutively.

The product designers indicate that it sometimes happens that an enquiry contains products that cannot be produced by the Dutch factory. The fact that these enquiries are sent to the Dutch production organization anyway indicates that the sales organizations are not very well acquainted with the production capabilities of the Dutch factory. Furthermore, it indicates that the Dutch project manager is not able to filter out every infeasible enquiry. The technical knowledge that is required to make a technical judgment of an enquiry is only available at the product design department. The decision what to do with an enquiry that contains products that cannot be manufactured, reject it or offer an alternative that can be produced, is made after consultation with the sales department. This consultation again is a time-consuming process, so it can stay unclear what happens to an enquiry for a long time.

The sales organizations and the Dutch production organization often disagree about the classification of enquiries. A rejection by the project manager or the product design department often leads to resistance at the sales organization. The English project manager has stated that he thinks the Dutch designers prefer PAs that are submitted by the Dutch sales organization, and therefore reject the English enquiries. The product designers indicate that they feel that the UK is throwing enquiries over the wall; forwarding them to the Netherlands without a proper first review.

4.4 DC 3: Resource control

This decision center is concerned with the deployment of resources to the different enquiries. Throughout every stage of the process it must be decided how many resources are assigned to the processing of the customer enquiry. This decision is especially relevant for the product design phase, as this is the most labor-intensive part of the quotation process. However, as Figure 3 indicates the decision of which resource to allocate to the enquiry is relevant throughout the whole process. Capacity use at the sales organization is not an issue as the tasks of the process that are performed there demand hardly any capacity: the first review of the enquiry and the sending of the final quote to the customer are tasks of limited size. The capacity issue becomes more urgent at the Dutch organization, where the enquiries are processed. As capacity at the product design department is limited, the resource control decision particularly affects the throughput time of the enquiries at that department.

The local sales organizations do not take the available capacity at the product design department into account when they forward the enquiries to the Netherlands. As they have no information about the current work load at the product design department, the decision to forward an enquiry to the Netherlands is made regardless of the available capacity.

(17)

17 process. In these cases he can reject the enquiry completely, or he can provide a cost estimate to the sales organization.

The last place where the resource control decision has to be made is the product design department itself. At that department two employees are responsible for the processing of all the enquiries, or PAs (Prijs Aanvragen). The two employees have to process many PAs next to the international project enquiries. Most of the requests can be regarded as regular work; these are relatively simple PAs that take less time to process than project enquiries. These PAs are mostly directed to customers in the wholesale market. Furthermore there are swap PAs and internal PAs. Swap PAs are related to the relocation of cable production to the Dutch factory. These are usually long-term projects, characterized by a relatively high complexity. The internal PAs are PAs that are not related to a customer request. They can include a request to design new products that are added to the standard product portfolio by product management, or requests for a new price of existing products as a consequence of changes in materials or routing.

Constantly there are a number of different PAs pending at the product design department. Therefore, the product designers must choose which request is the next in line to be processed when a previous PA is finished. Both employees have different methods of prioritizing their work. In the upcoming section the methods of prioritizing of both Employee 1 and 2 will be described.

Employee 1

Employee 1 focuses on the processing of the ongoing flow of regular PAs that arrive at the product design department. As can be seen in Figure 6, Employee 1 processes a large number of enquiries.

Figure 6: PAs processed by employee 1

Employee 1 has stated that his main personal objective is to limit the list of pending PAs at the department. He is aiming at not letting the number of open PAs exceed 20. In order to achieve this, Employee 1 in general chooses to process the relatively easy PAs that require less processing time, first. Table 1 supports this statement; the large majority of the PAs processed by Employee 1 are relatively simple regular PAs. In literature, this method of prioritizing is called the ‘shortest

Employee 1 466 44 94 51 0 50 100 150 200 250 300 350 400 450 500

(18)

18 processing time rule’ (SPT): select the job with the shortest imminent operation time (Panwalkar and Iskander, 1977). In some situations, however, the employee deviates from this rule. Sometimes the Dutch sales department asks the product designers to give high priority to a certain PA. Also in cases where the lead time of requests (in the employee’s opinion) tends to be too long, he will give priority to these waiting PAs.

This method of prioritizing results in longer throughput times for the international project PAs. As these PAs usually require more time of the product designer, the employee tends to postpone the processing of these PAs. Another reason why the relatively complicatedproject PAs have low priority is the perceived low chance of winning the project order after quotation. The “hit-rate” (percentage of quotations that results in a customer order) is low, less than 5%. This in combination with the large work load that comes with these enquiries causes that employee 1 often decides to process other PAs first.

Employee 2

The situation for Employee 2 differs from the situation of Employee 1: Employee 1 mainly processes the ongoing flow of regular PAs, while Employee 2 mainly processes a smaller number of large, complicated swap PAs (see Figure 7).

Figure 7: PAs processed by employee 2

A characteristic of swap PAs is that they usually contain a large number of constructions. This means that many different products need to be designed. The urgency of the different constructions of a PA is mainly dependent on the fact if there is a customer order placed for that particular cable or not. The PA lines with a customer order always get priority over the lines for which there is no customer order yet. Unless other priorities are indicated by the sales department, product management or higher management, the priority rule that is used for the processing of these customer orders is the First in – First out (FIFO) rule.

The number of customer orders for swap products is increasing. Therefore Employee 2 has been fully occupied with setting up constructions for customer orders during the period of this analysis. The

Employee 2 25 0 74 3 0 10 20 30 40 50 60 70 80

(19)

19 consequence of this is that only a few price enquiries could be processed by Employee 2 during the period of measurement.

The above shows that the long throughput time of the project enquiries is partly caused by the planning of tasks at the design department. The local sales organization does not take the available capacity into account when forwarding an enquiry. The project manager has a better insight in the work load at the product design department when an enquiry comes in. However, when an enquiry is sent to the product design department, other requests tend to be processed first. The current priority decisions at the product design department cause that the throughput time of project enquiries is longer.

4.5 DC 4: Identification of information for reuse

The identification of information for reuse is the final phase of the process. After a quote has been made the constructions of the new cables are added to Draka’s product database. When a customer requests the same cables again in the future, the constructions are already available so the product design phase can be skipped. By saving all the constructions that are designed, the product database continuously grows with new types of cables. When Draka participates more often in tenders for the international projects, the assortment of available constructions will become more and more comprehensive, which will help in future quotation processes. At the moment, Draka still experiences little advantage of the reuse of information. As the requested cables for the projects are usually very customer-specific, existing constructions of other cables do not help much in the design of these cables. Only when constructions for the customer are available, this can lead to significant time-saving in the design process.

Conclusion

(20)

20 Decision center Issue

1 Customization request initiation _• The process of gathering information on customer needs takes long

2 Classification _• Customer demands are still unclear in the classification phase

• Product design department receives requests for cables that cannot be designed in NL

• No agreement on the classification of enquires is reached

3 Resource control _• Project enquiries receive low priority at the product design department.

4 Identification of information for reuse _• Little of the information that is captured in the information system can be used in the quotation process for project enquiries due to the customer-specific nature of the requested cables

(21)

21

4.6 Recommendations for improving Draka’s responsiveness to international

project enquiries

The previous section has made clear that several factors throughout the quotation process affect Draka’s responsiveness. In this section, the possibilities of improvement of the process will be discussed, thereby applying the different factors that were found in the literature.

To what extent do the determinants of responsiveness affect the decision centers of the quotation process of Draka, and how can the determinants be used to improve the responsiveness?

It appeared that a number of the determinants of responsiveness (Reichhart and Holweg, 2007) affect the different decision centers of Draka’s quotation process. In the upcoming section, per determinant will be examined if and how the determinant can be used to improve the quotation process.

4.6.1 Demand anticipation

Reichhart and Holweg (2007) regard demand anticipation as one of the most obvious enablers for a supply chain to be responsive. The question is whether Draka can anticipate better on the actual demanded output.

Draka does not anticipate very well on the demand for international product quotes. Demand anticipation is impeded by the irregular flow of project enquiries, which makes it difficult to forecast the demand for projects in the upcoming period. Figure 8 shows the difference in arrival rate between the different PA types.

Figure 8: Overview of the arrival of PA lines during year 2009

The graph shows that while the arrival rate of regular PAs is rather smooth, the project PAs cause large fluctuations in the work load of the department. The irregularity of the arrival of project PAs

0 10 20 30 40 50 60 70 80 90 100 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51

(22)

22 makes it difficult to anticipate on the demand. Another hampering fact for the anticipation of demand is that the sales departments and the product design department are located in different countries. Therefore there is hardly any informal contact between the departments. The contact between the departments is mainly about enquiries that are already submitted by the customer. Because of this lack of contact, the product designers are not aware of which customers the sales organization is in contact with. In consequence, they cannot anticipate on any large enquiries that may result from these customer contacts. The project manager of the Dutch organization indicates that it is difficult to forecast the demand in large international projects. The sales organizations themselves do not have a well-grounded forecast of the demand, as the position in the international project market is still weak. When Draka is able to acquire a better position in the project market, it may be possible to make a reliable demand forecast.

However, even when a reliable forecast is available it will be difficult to anticipate on large demand fluctuations. The capacity of the product design department is not flexible. It is therefore difficult to temporarily expand the capacity in periods of high pressure. This makes it difficult to deal with large chunks of work. Due to the specific knowledge that is required to process the PAs, it is difficult to involve other people of the organization in the process. In the past the situation occurred that a retired employee of the product design department was asked to come back temporarily and help with the processing of a large project PA. The employee indicated that he will not come back again in the future, which means that the capacity to process PAs remains limited to two employees with occasional support of the manager of the product design department.

Concluding, demand anticipation in this situation is difficult for Draka but there is room for improvement. If Draka wants to improve its demand anticipation, the sales organizations should inform the product design departments as well as they can about upcoming project enquiries. Although the information about future requests may be limited, it can still be valuable to share this information. The product design department can better anticipate on future demand if they are informed about ongoing contact with the customer. Although capacity is fixed, the product designers can anticipate on future demand in other ways, for example by taking the future project enquiries into consideration when promising lead times in response to other requests.

4.6.2 Manufacturing flexibility

(23)

23 An approach to resolve this issue is to redesign the communication process so that less time is spent at this stage of the process. As a reminder, the communication process between the customers, the sales department and the product design department is represented again in Figure 9.

Figure 9: Representation of the communication process

As Figure 9 shows, in the current situation the product designers are not in direct contact with the customer. The fact that both the information request and the information delivery have to go passed the sales department slows the process down.

If the product designers would have direct contact with the customer regarding his demands, the representation of the communication process would look as shown in Figure 5.

Figure 10: Representation of the redesigned communication process

As shown in Figure 10, only four information flows are required when the product design department is in direct contact with the customer, instead of six in the current situation. This will reduce the time that is needed to get the exact customer demands clear. Furthermore, the information flows will be more efficient, because the product designers will be better able to communicate what exact technical information is missing.

(24)

24 unlikely that Draka will redesign the communication process as suggested. To improve the communication process without dramatically redesigning, the existing information flows should be speeded up. This means that on the one hand the product design department should minimize the time that passes before an information request is sent to the sales organization. On the other hand, the sales organization should make sure that information requests from the product designers are forwarded to the customer as soon as possible, and the information provided by the customer is directed to the product designers as soon as possible. This way of communicating should be driven by the people that are involved in the process, but can be supported by sending reminders regularly to the sales or product design department when a PA impends to become idle for too long.

Next to redesigning the communication process, it is also an option to reduce the need for additional contact between the product design department and the customer. To achieve this, the sales organizations should better review the enquiry before forwarding it to the Dutch organization. If obvious flaws in the enquiry, like incompleteness of the documents, are discovered earlier this will avoid the need for communication later in the process. However, a certain level of communication later in the process will always be required. The questions that arise during the thorough technical review of the enquiry by the design department cannot be prevented by the sales departments, as they are not trained to detect this kind of issues.

4.6.3 Coordination and resource sharing

The description of the quotation process showed that coordination is an important aspect of the quotation process, as multiple different organizations are involved. As is shown in section 4.3, coordination is especially required in the first two decision centers of the process; the customization request initiation and the classification of the enquiry. At this stage of the process, the enquiry must be directed to the right department. In practice, two problems occur: sometimes it takes too long before the enquiry is sent to the right country, and sometimes an enquiry is forwarded that the receiving production organization is unable to process. Both issues can be reduced when the sales persons that make the first decision regarding incoming enquiries get a better understanding of the capabilities of the different production facilities of the division. In order to do so, an up-to-date overview of the production capabilities should be available for the whole organization on the company’s intranet. Right now, such an overview is available but updating is lacking, which is critical given the large amount of production swaps that is going on at the moment.

4.6.4 Organizational integration

(25)

25 4.6.5 Spatial integration and logistics

The fact that the different departments involved in the quotation process are spatially dispersed has its effect on the performance of the process. The distance between the involved employees causes that there is no informal communication between the different departments, which has an effect on the process (see the section on Demand anticipation). Reichhart and Holweg however emphasize the logistical consequences of spatial integration. These consequences are not relevant for Draka, as the physical distance has no effect on the exchange of information over the internet and through phone calls.

4.6.6 Inventory, product architecture, information integration

The remaining three determinants for responsiveness are inventory, product architecture and information integration. Keeping an inventory is not possible in the quotation process. As the quotes are all unique and customer-specific, it is not possible to generate quotes in advance and keep them in stock. This determinant of responsiveness is therefore not relevant for the situation at Draka. Improving the product architecture in order to improve the quotation process is difficult because the cables that are requested must comply with very specific demands and standards. The information integration is related to the information processing part as described in section 4.6.2 Manufacturing flexibility. Therefore no recommendations regarding these three determinants are provided.

4.6.7 Mitigation of variety

Bramham et al. (2004) provide three different methods of mitigating variety in the quotation process. One of the proposed methods of mitigating variety is the filtering of requests for non-standard products. This method aims at reducing the number of non-non-standard requests. This method can be applied by the sales organization by trying to influence the customer and convince him to order standard products. At the moment, both the project manager in the Netherlands and the UK indicate that this filter is totally lacking. This has to do with the position that Draka currently holds on the project market. Because Draka is still a relatively small player, the sales organizations usually get involved quite late in the tender processes. At that moment, customer demands are already determined when Draka gets involved in the process, and it is hard to convince the customer that products of the standard product catalog might meet the customer’s wishes as well. This method of managing product variety in the quotation process can only be applied when Draka becomes more proactively involved in the tender processes of international projects. If the sales persons could direct customers more towards standard products or variants on standard products, this would reduce the pressure on the design departments and make it easier to respond to enquiries.

4.6.8 Absorption of variety

(26)

26 When competitors receive the same enquiry and they are capable of delivering the requested cables, Draka has a clear disadvantage. The sales organizations therefore prefer to comply with exact customer demands, as that increases the chance that the quote results in a sale. Another option is to make an estimate of the cost price of the cable, without already designing it in detail. This has a number of disadvantages. The estimator tends to at a safety margin to the estimate, in order to prevent offering the cable for a price that is too low. Furthermore, when an estimate is made, the customer cannot be offered the technical details of the cables. The sales managers indicate that customers prefer a complete tender, including all the specifications.

Another method of absorbing variety that is relevant in Draka’s situation is the processing of customer requests in order of importance, and the allocation of resources accordingly. This method relates to the procedure of prioritizing the PAs at the product design department. At the moment the PAs are prioritized mainly by the product designers and the Dutch project manager. This way of prioritizing causes that the international project enquiries have a relatively long lead time, as they generally get a low priority.

In the future, the product designers should not take commercial aspects into account when prioritizing their jobs. The product design manager indicates that the low perceived chance of winning the project causes that international project enquiries end up at the bottom of the pile. It is however not the responsibility of the product designers to make commercial decisions. Instead, the product designers should use a simple priority rule to prioritize their jobs. Panwalkar and Iskander (1976) provide an extensive overview of 113 possible scheduling rules. Most of the rules described are combinations or variants of the rules described in Table 7:

Priority Rule Description

FIFO – first in first out Process jobs in sequence of arrival

SPT – shortest processing time Process jobs with the shortest processing time first

LPT – longest processing time Process jobs with the longest processing time first

EDD – earliest due date Process jobs with the nearest due date first ST – slack time Process jobs with the least amount of slack first

(available time before due date – time remaining for the operations)

Value Select jobs with the highest value first

Table 7: Overview of priority rules

(27)

27 situation this could be a suitable planning method, as both due dates and processing times are important variables.

The two PA characteristics that the product designers should know to be able to apply the slack time rule are:

• PA due date

• Processing time of the PA

The due date of PAs should be clearly communicated by the sales organizations. The processing time of the PA should be estimated by the product designers, as processing times are not fixed.

It may happen that multiple PAs have an equal small (or even negative) slack. This means that both PAs should be processed immediately, in order to meet the customer’s due date. When this is the case, the commercial importance should determine which PA must be processed first. However, the definition of commercial importance of the different customer requests differs between the different organizations. A number of factors determine the relative commercial importance of PAs, such as:

• Potential profit margin

• Probability of winning the order • The relation with the customer

As the opinions about the commercial factors of importance differ, the question is: who should determine the importance of the PAs, and thus the sequence of processing? Because it is a commercial consideration, the decision cannot be made by employees of the design department as it is not the responsibility of the product designers to make commercial decisions. The product designers will tend to choose the less complex PAs first, as is shown in section 4.4, which is not necessarily the best option.

Leaving the decision to the sales organizations that submit the enquiries will also lead to problems. They cannot make a fair comparison between the enquiries that are initiated by them and those initiated by another sales organization. Logically, a sales person tends to assign the highest commercial importance to the enquiries that he initiated.

(28)

28

5. Conclusion and discussion

In this section, the main research question will be answered. The main research question sounded: Which factors influence the quotation process of Draka E&I Europe, and how can these factors be changed so the responsiveness to customer enquiries increases?

It appeared that a number of factors influence the quotation process, which affects the responsiveness to customer enquiries.

The lack of demand anticipation is a major issue in the quotation process. The project enquiries arrive irregularly at the product design department, and cause a large peak in the work load. This causes difficulties for the product design department to process these requests. It is however difficult to resolve this issue. The first reason is that the sales organizations are not able to forecast the demand. The second reason it is not easy to adjust the available capacity at the product design department to deal with peak demand, as specific technical knowledge is demanded. An improvement of the demand anticipation can be achieved when sales organizations share important information regarding possible future enquiries with the product design department. If they do so, the product designers may be able to take the project into consideration when accepting other requests.

The manufacturing flexibility is low, due to the long throughput times of enquiries. The long throughput times are partly caused by the extensive communication process between product design, sales and the customer. Redesigning the communication process by allowing direct contact between product design and the customer would reduce the length of the process. This is however contrary to the philosophy of the division, to provide local contacts for the customer. Furthermore, the product designers are not trained in contact with the customer. The communication process can be shortened when product design and sales minimize the time that they use to forward a request. Making reminders can help in achieving this.

The communication process can also be shortened by reducing the need for communication during the design process. This can be achieved by improving the first filter of the enquiry by the sales organizations.

Mitigating variety by the filtering of non-standard requests is a potential area of improvement for Draka, but it is not easily achieved. Draka needs to obtain such a market position that it gets involved in talks about the large projects earlier in the process. Only then will it be possible to direct the customer to demand standard products.

Absorption of variety by offering the closest appropriate standard product to the customer would increase Draka’s responsiveness greatly. Applying this method would reduce the need for communication on the exact customer demands, and eliminate the time spent on the design of the constructions from the lead time. The largest drawback of this method is that the chance of winning the order will reduce substantially.

(29)

(30)

30

References

Bramham, J., MacCarthy, B. and Guinery, J. 2004. Managing product variety in quotation processes. Journal of Manufacturing Technology Management, 16(4): 411-431

Chen, H., Daugherty, P.J. and Roath, A.S. 2009. Defining and operationalizing supply chain process integration. Journal of Business Logistics, Vol. 30, No. 1

De Treville, S., Shapiro, R.D. and Hameri, A.P. 2004. From supply chain to demand chain: the role of lead time reduction in improving demand chain performance. Journal of Operations Management, 21(6): 613-627

Forza, C. and Salvador, F. 2002. Managing for variety in the order acquisition and fulfilment process: The contribution of product configuration systems. International Journal of Production Economics, 76(1): 87-98

Hartmann, W., Fischer, A. & Nyhuis, P. 2009. The Impact of Priority Rules on Logistic Objectives: Modeling with the Logistic Operating Curves. Proceedings of the World Congress on Engineering and Computer Science

Hvam, L., Simon, P. and Nielsen, M.K. 2006. Improving the quotation process with product configuration. Computers in Industry, 57(7): 607-621

Panwalkar, S.S. and Iskander, W. 1977. A Survey of Scheduling Rules. Operations Research. 25(1): 45-61

Reichhart, H. and Holweg, M. 2007. Creating the customer-responsive supply chain: a reconciliation of concepts. International Journal of Operations and Product Management, 27(11): 1144-1172 Slack, N., Chambers, S. and Johnston, R. 2004. Operations Management. Prentice Hall.

Winning More Projects