ORGANIZATION OF THE FOOD PROCESSING INDUSTRY AND ITS IMPACT ON SUPPLY CHAIN COMPLEXITY: A MULTIPLE CASE STUDY

(1)

ORGANIZATION OF THE FOOD PROCESSING INDUSTRY

AND ITS IMPACT ON SUPPLY CHAIN COMPLEXITY: A

MULTIPLE CASE STUDY

University of Groningen Faculty of Economics and Business

Department of Operations

Postbus 800, 9700 AV Groningen, the Netherlands

Final Thesis

ARJEN VAN DER MEER [S2593246]

MSc Thesis SCM 2015

(Dr. D.P. Van Donk & H. Dittfeld)

(2)

2

Acknowledgements

Firstly, I’d like to thanks Prof. Dr. Dirk-Pieter van Donk and Hendryk Dittfeld. They have guided me through this process with their feedback and constructive criticism, resulting in the work that now lays before you. Also, many thanks to all the organizations that were willing to participate in our research. Their openness, knowledge and enthusiasm where much appreciated and have had a big influence on the motivation for this project. In accordance, I’d like to thank my fellow students in the ‘SC-complexity within the FPI’ group for their effort, feedback and ideas and in particular Tom Grevers for our lengthy discussion sessions during the final stages of thesis writing.

(3)

3

Abstract

Purpose: This paper investigates how the organizational structure of organizations within the food processing industry drives supply chain complexity. The organizational structure is defined through an aggregated model by James & Jones (1976) and its influence is measured on drivers of complexity as identified by De Leeuw, Grotenhuis & Goor (2013). In addition, several characteristics of the food processing industry are taken into account as identified by Van Donk (2001) to create a sector-specific understanding.

Design/ Methodology/ Approach: Ten organizations that operate within the food processing industry are part of a multiple case study that is conducted through interviewing various managers on both a company and operational level. These interviews together with data from a survey are used to group cases in order to find patterns between similar and different groups. All cases are subject to a multi-step coding procedure to identify the constructs as defined in this research.

Findings. The data analysis shows no sign of the organizational structure influencing supply chain complexity within the food processing industry. However, the influence of food processing industry characteristics on both the organizational structure and SC complexity is identified. In addition, some differences are found between organizations based on their food processing industry characteristics.

Theoretical Implications. The insights show the need for a sector specific measurement tool for supply chain complexity. In addition, the apparent link between the characteristics of the food processing industry and both the organizational structure and the experienced supply chain complexity drivers calls for additional research.

Originality/ Value. This paper contributes to an understanding of how the organizational structure drives supply chain complexity within the food processing industry. Further, this paper provides insight in how specific characteristics of the food processing industry influence the perceived supply chain complexity.

(4)

4

Introduction

An organizational structure is often designed by top-management or gradually evolved over the years to best fit the business strategy. Nevertheless, the organizational structure is mentioned by Matt (2007) as an important driver to address upon mitigating supply chain (SC) complexity and therefore making it an interesting topic to investigate. The relevance of this research links to the loss of performance due to an increase in SC complexity (Perona & Miragliotta, 2004) and the necessity to decide on appropriate actions to deal with SC complexity (De Leeuw, Grotenhuis, & Goor, 2013). However, SC complexity is a rather new research field and although researchers have already been proposing methods to measure and identify possible SC complexity drivers, research is lacking in terms of the actual influence and manageability of these drivers on SC complexity (Manuj & Sahin, 2011). This paper uses drivers from a recent model by De Leeuw et al. (2013) to indicate the level of SC complexity, these drivers include: uncertainty, diversity,

size, variability, structure, lack of cooperation, speed and lack of information synchronization. In

addition, the researchers opted to further investigate the organizational structure opposed to other SC complexity drivers, as organizational factors are key determents for a firms’ performance (Hansen & Wernerfelt, 1989).

In accordance with the specific driver choice, it was further decided to investigate a particular sector. Therefore, due to the lack of sector-specific research, but mostly due to a change in customer wishes showing an increase in SC complexity, the researchers have decided to investigate the food processing industry (FPI). These changes in customer wishes are indicated by Van Donk (2001) and include an increase in product variety, shorter lead-times, smaller and more frequent batches, and a longer shelf-life. In addition, several specific characteristics of FPI, such as perishability (Georgiadis, Vlachos, & Iakovou, 2005) and risk of contamination (Mahalik & Nambiar, 2010) have increased planning difficulty (Van Wezel, Van Donk, & Gaalman, 2006) and hence might influence SC complexity. However, these specific characteristics have yet to be linked to the complexity drivers as identified in the model by De Leeuw et al. (2013). Therefore, this research will gain insight in how the structure of organizations, which operate within the FPI, influences SC complexity. In accordance, managers and researchers alike will be able to better manage and analyse the actual influence of complexity drivers within FPI.

Thus far, research has mostly been evolving around measuring SC complexity and identifying possible SC complexity drivers (Manuj & Sahin, 2011). However, due to the lack of an easily employable quantitative measure for SC complexity (Jacobs, 2013) researchers are still unable to make the actual influence and manageability of the drivers apparent to managers (Manuj & Sahin, 2011). Consequently, this grey area within SC complexity studies results in a gap in literature. Therefore, this research will assess the role of the organizational structure on SC complexity within the FPI in order to identify its actual influence and manageability, resulting in the following research question:

How does an organization’s structure influence the supply chain complexity within the food processing industry?

(7)

(8)

8

Theoretical Background

The following section is dedicated to the understanding of key terms and their underlying relationship. Therefore, the paper will begin by briefly elaborating on the characteristics of the food processing industry and supply chain complexity. Additionally, the middle section will give a rather in-depth review of the organization structure. Consequently, the final section will link the several variables through a theoretical framework based around a conceptual model in order to establish possible relationships.

2.1 Characteristics of the food processing industry

The food processing industry is different from other processing companies due to its distinctive perishable characteristics (Van Donk, 2001). In addition, changes in customer wishes (Van Donk, 2001) with regards to more product variety, shorter lead-times, smaller and more frequent batches and a longer shelf life have resulted in an increase in planning difficulty (Van Wezel et al., 2006). Furthermore, the use of expensive single-purpose plants, long and sequence-dependent set-up times, variability in supply and quality of materials and variable yield (Van Donk, 2001) together with the use of intermediate storage, an added chance of contamination (Kilic, Akkerman, Donk, & Grunow, 2013) and rules and regulations regarding hygiene and product contamination (Trienekens & Zuurbier, 2008) all add to an increase in SC complexity within FPI. Consequently, given the fact that SC complexity influences a firm’s performance (Bozarth, Warsing, Flynn, & Flynn, 2009) it becomes highly apparent to mitigate SC complexity within FPI as the industry deals with minimal profit margins (Kilic et al., 2011; Maloni & Brown, 2006). Additional information on SC complexity will be given in the next section of the paper.

2.2 Supply chain complexity

Rapid innovation (Perona & Miragliotta, 2004), increasingly demanding customers (Jacobs, 2013) and globalization of supply chains (Manuj & Sahin, 2011) have all contributed to a gradual increase in SC complexity during recent years. Additionally, planning activities have become increasingly complex due to a wider product variety, smaller batches, shorter-lead times and different tiers within an integrated supply chain (Jacobs, 2013). Furthermore, the globalization of the supply chain has had organizations deal with cultural differences, language barriers, trust, quality issues and lack of knowledge (Nayak & Taylor, 2009). However, despite Bozarth el al. (2009) indicating the negative influence of SC complexity on a firm’s performance, research is still lacking a thorough understanding of the drivers behind SC complexity (De Leeuw et al., 2013). The above mentioned lack of understanding results in the lack of an uniform measurement tool for SC complexity (Jacobs, 2013). In turn, researchers are unable to translate their findings in useful management approaches (Manuj & Sahin, 2011).

This research will adapt SC complexity drivers from a recent model by De Leeuw et al. (2013) that identifies eight drivers for SC complexity: uncertainty, diversity, size, variability, structure,

speed, lack of information synchronization and lack of cooperation. However, other agents in the

(9)

9

Driver Definition Key Sources

Uncertainty The lack of predictability and reliability of demand and of supply chain in processes

Diversity The number of different suppliers and customers, the level of customization of this range and the range of different products /services offered

Frizelle and Woodcock (1995), Frizelle (2004), Funk (1995), Isik (2010), Perona and Miragliotta (2004), Sivadasan et al. (2004), Sivadasan et al. (2006), Zhou (2002) Size Relative number of volume of products or

activities, including batch sizes for purchase, production, supplies

Bozarth et al. (2009), Frizelle (2004), Funk (1995), Isik (2010), Perona and Miragliotta (2004), Vachon and Klassen (2002)

Variability Sudden, large and variable changes in requirements over time

Bozarth et al. (2009), Isik (2010), Sivadasan et al. (1999)

Speed Required responsiveness across the supply chain in terms of throughput times, delivery times and frequencies

Bozarth et al. (2009), Sivadasan et al. (2004)

Lack of information synchronization

The lack of coordination and control of information on requirements and resources over time

Sivadasan et al. (1999), Zhou (2002)

Table 1 – SC complexity drivers

DeCanio et al. (2000) and Hansen & Wernerfelt (1989) have identified the organizational structure as the main force behind a firm’s performance. In addition, Matt (2007) has identified the simplification of the organizational structure as the first step towards mitigating the influence of SC complexity. The next section of the report will elaborate on the organizational structure and its variables.

2.3 Organizational structure

The dimensions of an organizational structure have been defined through an extensive literature research by James & Jones (1976) as being: (1) organization size, (2) centralization of decision making and authority, (3) configuration, (4) formalization, (5) specialization, (6) standardization, and (7) the interdependence of organizational components. However, for this paper, the researchers have opted to merge categories one through three, and four through six together for showing similarities. In addition, new literature has been found to strengthen the merged dimensions. The newly comprised model is illustrated in table 2.

(10)

10 The model above shows the organizational structure as consisting of three independent dimensions comprising of several variables. The first dimension, the configuration of an organization, includes the size of the organization (James & Jones, 1976); geographical dispersion (Pertusa-Ortega, Zaragoza-Sáez, & Claver-Cortés, 2010); number of hierarchical levels (De Leeuw et al., 2013; Pertusa-Ortega et al., 2010) and the number of processes and processing steps (all steps that add value to the final product within the organization’s plant) (De Leeuw et al., 2013). The second dimension is labelled formalization. The formalization of the organizational structure refers to the number of rules and regulations (Pertusa-Ortega et al., 2010) and the extent to which processes and the decision making are standardized through the use of procedures (Chen & Huang, 2007). The third dimension, vertical integration, has been related to research by Sivadasan, Efstathiou, Calinescu & Huatuco (2006) and relates to the extent to which various subdivisions of an organization work interrelated and share knowledge between departments (Chen & Huang, 2007). These three dimensions are expected to cover all organizational factors that prove to be key determents for a firm’s performance (Hansen & Wernerfelt, 1989) and therefore possibly relate to the level of SC complexity as experienced by organizations. The next sections will further elaborate on the three identified dimensions of the organizational structure and their accompanying variables.

2.3.1 Configuration

The way that an organization is configured could have an influence on the level of SC complexity. The dimension’s variables will be enhanced below, ensuring an understanding on which possible relationships can be built in the theoretical framework at the end of this chapter. Organizational size – The size of an organization as identified by James & Jones (1976) often relates to the total assets and/or revenue of an organization (Gallo & Christensen, 2011) or the number of employees within an organization (Chen & Huang, 2007).

Geographical dispersion – The number of facilities an organization has and the number of countries an organization is active in, together with the distance between these facilities all relate to the geographical dispersion (James & Jones, 1976). In addition, Pertusa-Ortega et al. (2010) have identified geographical dispersion as a dimension of structural complexity. This relates to the erection of obstacles (centralization) and efficiency improvements (decentralization) as indicated by Brammer & Millington (2004). In addition, it is stated by Brickley and Dark (1987) as cited in Michael (2000) that geographical dispersion negatively affects quality.

Number of Hierarchical levels - The number of hierarchical levels in an organization can have either a positive or negative influence on the effectiveness of the decision making process (Zhou, 2002). For instance, when the number of layers within the organization becomes larger, the time to market for new products increases (Lam & Chin, 2005) and it becomes more difficult to contract new suppliers (Aissaoui, Haouari, & Hassini, 2007). However, when procedures are regulated by rules, steeper hierarchical structures do not necessary negatively affect a firm’s performance (Anderson & Brown, 2010).

(11)

11 steps include all the steps a product goes through, within the factory, to be processed into the final product. Frizelle & Woodcock (1995) argue that by increasing the number of processes and processing steps a product has to go through, it becomes harder for organizations to adequately react to changes in demand.

2.3.2 Formalization

This dimension of the organizational structure is closely related to employee behaviour. The behavioural aspect within formalization is not related to culture but to rules and regulations related to day-to-day workings of the organization. Another aspect of formalization is

standardization or the way in which processes (e.g. production, new product development

and intake of raw materials) are shaped through rules and procedures.

Rules and regulations - Rules and regulations are defined by Pertusa-Ortega et al. (2010) as the degree to which decision making and working relationships are governed by rules, procedures and standard policies. Pertusa-Ortega et al. (2010, p. 312) further state that implementation of rules and regulations “limit the chances for organization members to communicate and interact

with one another (López et al., 2006)”. These rules and regulations are seen as a limitation as

they limit the flexibility of the organization and its drive for innovation. In addition, adhering to rules and regulations make day-to-day business more complex. However, other researchers also argue that rules and regulation can make for structured knowledge gathering and therefore enable the organization to make quick decisions in time of distress (Kern, 2006).

Standardization - Standardization is defined as the degree to which processes are made homogeneous through procedures and the extent to which employee behavior adheres to these procedures (Chen & Huang, 2007). However, this would also mean obstructing employees’ willingness to discuss and consider alternatives and hence negatively influence innovation (Chen & Huang, 2007). Nevertheless, Choi, Dooley & Rungtusanatham (2001) further argue that too much innovation can undermine managerial predictability and work routines. Therefore, a balance should be found between control and room for innovation.

2.3.3 Vertical Integration

The reason to vertically integrate within the supply chain can be based upon assessing an organization using the transaction cost theory (Acemoglu, Griffith, Aghion, & Zilibotti, 2010; Garcia Martinez, Poole, Skinner, Illés, & Lehota, 2006; Karantininis, Sauer, & Furtan, 2010). This theory assumes that a company should opt for vertical integration when “there is greater

specificity and holdup is more costly, and that vertical integration should enhance investments by all contracting parties (Acemoglu et al., 2010, p. 990). Vertical integration adds to supply chain

(12)

12

2.4 Research Framework

The abovementioned research translates in the conceptual model as shown below (Figure 1), this model illustrates the expected relationship between the various components of the organizational structure and SC complexity. Moreover, the influence of the FPI characteristics on both the organizational structure and SC complexity will be tested together with its moderating role on the relationship. The understanding of the model will answer the research questions and will give researchers and businesses alike new insights in how to manage SC complexity within the FPI.

(13)

13

Methodology

The aim of this research is to gain insight in the role of an organization’s structure on SC complexity within the FPI. The lack of previous research in regard to the intended aim makes this an exploratory research. Therefore, the multiple-case approach was deemed most appropriate as it enables for rich data from real-life cases (Yin, 2013) which cannot be achieved through conducting a mainly statistical analysis (Meredith, 2002). In addition, a case study approach lends itself for objectivity through the use of “how” and “why” questions. This is mainly achieved through semi-structured interviews, which are used as a guide but also allow for additional information gathering. Furthermore, the approach can be used for cross-case analysis (Yin, 2013), making the results more generalizable than those of a single case approach (Eisenhardt & Graebner, 2007).

3.1 Unit of Analysis

The plant (within FPI) at which we want to determine the influence of the organizational structure on the SC complexity can be regarded to as the unit of analysis. Due to confidentiality reasons the plants will be identified with a letter ranging from ‘plant A’ to ‘plant L’. The scope of the paper entails both the operational level and company level. This scope enables for a full understanding of both the organizational structure and the level of SC complexity within the facility.

3.2 Data collection

The collection of data is done through an interview protocol (qualitative) and accompanying questionnaire (quantitative) embedding both semi-open and closed questions. The protocol will further be used to guide the interviewers and enables them to touch on all relevant subjects. In addition, collection of the data is done in teams of two. This will enable for a division of tasks, where one researcher conducts the actual interview and the other takes notes. Furthermore, the use of teams enables for a broader perspective and more creativity to be put towards conducting the interview. Interviews were preferably done with two managers to ensure full visibility of both a company level (Supply chain, Sales or Marketing manager) and operational level (Production planner or Head of Operations), these interviews took approximately 45-90 minutes per respondent. Practice has further proven that discussion and feedback between the researchers after each interview resulted in improvements regarding the questionnaire, interview protocol and probing techniques.

3.3 Development of interview protocol

(14)

14

3.4 Operationalization of the organizational structure (configuration)

The construct of the organizational structure is proposed in an aggregated state based on a literature review by James & Jones (1976). In this aggregated state the construct deals with three dimensions of which the configuration is measured based on quantitative data derived from both the interview and the accompanying survey. The other constructs are measured through coding which is explained later on in the chapter. The various components of the configuration will be measured using pre-defined scales. Where, for example, the number of employees is measured as {<100 is small}, {100 – 399 is medium} and {400 >, as large}. The components together make up the dimension and hence indicate the organization’s configuration.

Table 3 – Organizational structure configuration measurement

3.5 Sample selection

The data pool is comprised of 10 organizations (table 4) that are active within the FPI. The organizations are selected based on the FPI characteristics as identified by Van Donk (2001) as having expensive single-purpose plants and working with perishable materials. In addition, organization are to employ managers on both a company and operational level. Consequently, theoretical replication is attained through having an array of different organizations in terms of their organizational structure. In addition, literal replication is achieved through analyzing organizations with similar organizational structures based on their level of SC complexity. The sample is depicted below, the names are changed to “Plant A” through “Plant J” for confidentiality reasons. The organizational structure is identified through the size, number for full-time employees (FTE), the volume and number of facilities. The sample fits the purpose of the study as the organizations show both similarities and differences making it suitable to group organizations and make comparisons amongst groups.

Plant Size Main product # FTE Volume Facilities

Plant A Large corporation Potato starch 900 550.000 tons 6

Plant B Medium organization Infant nutrition 500 20.000.000 cans 5

Plant C Small organization Energy drinks 21 15.000.000 cans 1

Plant D Large corporation Infant nutrition 800 130.000 tons 7

Plant E Medium corporation Infant nutrition 225 52.000 tons 7

Plant F Small organization Flower 110 200.000 tons 1

Plant G Large corporation Potato products 105 100.000 tons 15

Plant H Medium organization Chicken products 400 364.000.000 chicks 3

Plant I Small organization Pet snacks 230 28.000 tons 3

Plant J Medium organization Industrial bakery 500 50.000 tons 11

Table 4 – Sample

Dimension _Measurements _{Key Sources}

Configuration

Size of the organization # employees Chen & Huang (2007) revenue (€) Gallo & Christensen (2011) product volume De leeuw et al. (2013) batch size

Geographical dispersion # facilities James & Jones (1976) avrg. km between facilities

# countries

Number of hierarchical levels # of hierarchical levels Zhou (2002)

(15)

15

3.6 Quality of Research

Karlsson (2009) has identified internal validity, external validity, construct validity and reliability as the most important criteria when conducting case research. Validity in general is used to ensure that what needs to be measured is actually measured. Internal validity was achieved through pattern matching, these patterns where found in both FPI characteristics and the organizational structure. External validity is difficult for SC complexity as researchers have not yet agreed upon a uniform measurement tool (Manuj & Sahin, 2011). However, the model by De Leeuw et al. (2013) is very recent and embedded in literature and therefore adds to external validity. In addition, the organizational structure is well embedded in literature and therefore external validity is ensured for this construct. Furthermore, construct validity is embedded in data source triangulation through the use of interviews, a survey, documents and databases for the collection of data. In addition, validity was further enhanced by multiple sources of evidence (operational level and company level managers). The reliability of the research is ensured by achieving a chain of evidence. This chain consists of a structured interview protocol, the recording of the interview and the transcribing of the interview to ensure reliability.

3.7 Data analysis

The data analysis, given the exploratory nature of the research, evolves around finding patterns between cases in order to develop new theory (Yin, 2009). The data analysis was started by doing a within-case analysis, which was followed by a cross-case analysis. This cross-case analysis helps in comparing, generalizing and gaining a thorough understanding of the within-case findings. Each case was analysed on their FPI characteristics, their experience with SC complexity drivers and finally on their organizational structure (Appendix 1 through 3).

(16)

16

Results

The first section will provide an overview of the FPI characteristics that have been mentioned throughout the interviews. This representation is used in the second section, where the influence of the SC complexity drivers are identified and linked to the FPI characteristics. The third section will show a break-down of the three organizational constructs: configuration, formalization and the level of vertical integration and their link to SC complexity and the influence of the FPI characteristics on this relationship.

4.1 FPI characteristics

The interviews have been analysed on the FPI characteristics as previously identified. This has resulted in an overview as presented in table 5 where perishability, food safety & regulations and

changes in customer wishes have been mentioned the most. Furthermore, it can be noted that

supply and quality variability is often mentioned by organizations that deal with a natural products e.g. grain “last year we got a harvest with a very fluctuating quality” (Purchasing manager, Plant F) and “Potatoes are a natural product. In our process it is the biggest variable” (Operations manager, Plant G). However, organizations that deal with the natural product milk indicate that “we have continuous process controls and we see almost no difference” (Operations manager, Plant E) and “The raw materials have to meet requirements, so we make sure that the

quality is above a certain minimum” (Supply chain manager, Plant D). Indicating an interesting

difference within the sample based on how the supply of materials is managed.

Table 5 – FPI Characteristics mentioned in sample

The organizations are found to be highly subject to changes in regulations “as food safety

regulations became more and more demanding” (Head of operations, Plant C). These strict rules

make it that “{Plant A} has suppliers state that they do not supply to you because you do not meet

their safety requirements” (Purchasing manager). In another plant the regulations “are almost as high as the pharmaceutical industry” (Supply chain manager, Plant D) making “the handling of the different food safety regulation more complicated than the actual production” (Head of

operations, Plant C). Moreover, through the increase in customer wishes, organizations are found to have an increasing array of small products which have a big influence on their experienced level of SC complexity. One manager went as far as to say: “I have been here since February, the

first idea I had when I saw all the articles: “let’s get rid of the bottom 20%” and then we’ll see what happens” (Sourcing manager, Plant F). In addition, the Planning & distribution manager of

plant H indicates to “produce and deliver 80 different kinds of chicken products. However, “many supermarkets have their own private label with a specific dish and logo” increasing their production to “around 300 different products a day”.

FPI Characteristics Plant A B C D E F G H I J # %

Perishability 1 3 3 1 1 3 2 2 1 1 18 23%

Long and sequence-dependent processes 4 3 1 1 1 2 12 15%

Supply variability 2 1 1 1 1 1 7 9%

Quality variability 1 1 1 1 2 1 7 9%

Food safety and regulations 2 3 6 1 1 2 1 16 20%

Changes in customer wishes 5 1 4 2 1 2 1 3 1 20 25%

(17)

17

4.2 SC complexity within FPI

The first finding relates to the differences in SC complexity (table 6) and in particular to the influence of FPI characteristics on this variation within the sample. It is interesting to note that the organizations with the highest SC complexity (plant A, F, G and H) all deal with natural materials. This product relates to characteristics of the FPI where there are great varieties in yield and in the quality of the raw materials. This is also indicated by the managers of the plants, where the planning & distribution manager of plant A stated that: “if the sun is shining and it will rain,

we will get a lot of potatoes and thus starch. If it doesn’t rain and there won’t be any sunshine, at the end we have got very little starch” and plant G, who also deals in potatoes, mentioned: “With potatoes it is about the fact that it is a natural product. This makes it complex, as not one year is the same” (Logistical manager). In addition, the purchasing manager of plant F further indicates

that as “You buy it beforehand, you pay a certain price for a batch, and you receive that batch

later” which adds to your uncertainty, as it is unknown what will be delivered and you want to be

“ensuring a continuous quality”. The organization with the highest measured complexity in this group, Plant H, also indicates “to deal with a heterogeneous product at the slaughterhouse” as

“a chicken is a living being and grows on its own manner”. In addition, it has to cope with “short delivery lead times and the external influences, like the weather” and the fact that their

location “produces around 300 different products a day”.

Diversity Uncertainty Size Variability Speed

Lack of information synchronization Overall SC complexity score Plant A ++ + + +- ++ - + Plant B ++ ++ - + - - +- Plant C - + - + - - - Plant D + - - + - + - Plant E +- + + +- -- +- +- Plant F ++ ++ - + + - + Plant G +- + +- ++ + - + Plant H ++ ++ + + ++ +- ++ Plant I +- + n.a. + +- - +- Plant J + +- + - + - +-

Table 6 – Impact of SC complexity drivers per organization

The organizations that score on a mid-level when it comes to the SC complexity drivers, greatly vary in product type. Whereas, two organizations produce infant nutrition (Plant B and E), one produces dog treats (plant I) and one is an industrial bakery (plant J). These plants have in common that they have a small number and stable supply of raw materials. Moreover, their similarity in SC complexity is found in their diversity and in particular in the number of different packages they are required to offer. This further relates to the change in customer wishes within the FPI. The various managers establish this change by saying “There are customers who want it

in big bags, and there are customers who want it in 25kg bags, the other want such a label and the other wants such a specification” (Operations manager, Plant E) or “Our packaging materials are even more with around 1000 different forms as every customer requires its own labels”

(Head of Operations, Plant I). Another manager indicates that they “have 300 different types of

(18)

18 products has to deal with the fact that “each type contains many kinds of different product.

Therefore, we have hundreds of final products” (Quality manager, Plant J).

The final two plants show a limited impact regarding the SC complexity drivers. The first plant is part of a big multinational active in the dairy industry. However, “{location censored} is a

production location” and therefore not very influenced by the drivers, as the supply chain

manager also indicated “I look forward for 18 months and know already approximately what we

are going to be producing the coming year”. In addition, “There is a very strict organization within the cooperation, who go to the farmers to tell them what they have to do” (Supply chain

manager, Plant D) as to ensure high and steady quality. This collaboration and the single-purpose characteristic of this plant limits its level of SC complexity. The other plant is a small company from Germany that produces energy drinks. However, it differs from the other plants in the sample as it has outsourced its processing activities as it “works together with a single raw

material supplier as well as co-packer” (Head of operations, Plant C). The plant has mitigated

the influence of its SC complexity drivers through their supplier and co-packer, who “are able to

absorb peaks in demand which is contractually guaranteed to us and leads to contractual penalties if they cannot fulfil agreements” (Head of operations, Plant C). It works with

production slots that are reserved six weeks ahead of time, leaving the organization enough time

“to procure materials ahead-of-schedule”.

4.3 Organizational structure

Upon investigating the configuration of the organizations in the sample (Table 7), we have come across a variety of different organizations. However, after assessing the organizations it was decided to neglect the revenue as one manager stated: “Money is totally not relevant … our

turnover is fully depended on the cost of raw materials” (Purchasing manager, Plant G). In

addition, only a minor part of the respondents were willing to share their monetary revenue with us. Therefore, this research will use the yearly volume to indicate an organization’s size. Moreover, the number of processes was also neglected as the plants all had either one or two processes, having no influence on the final representation.

Table 7 – Configuration of the sample

4.3.1 Configuration vs. SC Complexity

The first category encompasses small organizations, which includes three companies as illustrated in table 8. In this category we find organizations with a low number of employees, often one facility, they are active in a small number of countries and have a limited amount of hierarchical levels. “It is a fairly small organization, a company where you have way more

authority than when you work for a bigger concern … and short lines, that makes it easier”

# FTE Volume Batch size Facilities

Distance between facilities Countries Hierarchical levels Processing steps CONFIGURATION

Plant A 900 550.000 tons 200 tons 6 600 km 6 5.5 10 LARGE

Plant B 500 20 mil cans 12.000 SKU 5 1500 km 3 7 6 MEDIUM

Plant C 21 15 mil cans 825.000 cans 1 0km 10 3 5 SMALL

Plant D 800 130.000 tons 350 tons 7 250km 15 6 6 LARGE

Plant E 225 52.000 tons 85tons 7 250km 32 3.5 6 MEDIUM

Plant F 110 200.000 tons 75 tons 1 0 km 2 3 9 SMALL

Plant G 105 100.000 tons 550 tons 15 1500km 4 3 14 LARGE

Plant H 400 364 mil chickens 25 tons 3 150km 13 3.5 7 MEDIUM

Plant I 230 28.000 tons 900kg/h 3 700 km 2 3.5 3 SMALL

(19)

19 (Logistical planner, Plant F). The small organizations show dissimilarities in the extent to which they are influenced by SC complexity. This difference is found in the type of products that are produced. As mentioned before, Plant F processes a natural product, highly increasing its SC complexity through variability in both harvest and quality. However, they also seem to lack the expertise of bigger organizations in coping with SC complexity “We do not have an ERP or data

system that knows when what customer orders what product … like a {censored} that uses SAP

(logistical planner, Plant F) further increasing its high level of SC complexity. However, through their low level of hierarchical levels “Many things within {Plant F} are done together, many

decisions are made together, this is necessary if you want to make your margins on a deal. So it is basically, working outside of the pillars, working together as much as possible is very important for a flower factory in general” (Purchasing manager, Plant F) they are able to cope

with SC complexity.

Plant C scores the lowest on complexity as it has outsourced its production activities to a co-processor, which seems to be the norm for the beverage industry. The only real uncertainty indicated by the head of operations is that they “only work together with a single raw material

supplier as well as co-packer” where “In case of an unexpected event, like a breakdown of the filling machine or a supplier does not deliver the needed materials, we suddenly face major problems”. However, these problems can be quickly resolved through their short lines and the

fact that “we work closely together and exchange information” (Head of operations, Plant C). The final plant in this group, Plant I, differs from the other two in the fact that it produces dog treats for which regulations are less strict “these raw materials are of category three and not

usable for human beings anymore” (Head of operations, Plant I). Its SC complexity stems from

its market being highly volatile and orchestrated by discounters who “want to receive their goods

real quick and on time” and “ask for new articles to offer” resulting in “huge variety of forms and colours” (Head of operations, Plant I). They are capable of dealing with this level of SC

complexity through their limited number of hierarchical levels, resulting in their “communication

within the company functioning relatively trouble-free”.

The second category, medium organizations, is comprised of four companies that have an average number of employees, a limited number of facilities and both an average number of hierarchical levels and processing steps. Three plants are coping rather well with SC complexity, while one scores the highest out of the sample. This plant’s planning & distribution manager is quoted saying: “We operate in a very complex market and chain, complexity is an opportunity for

us to excel as one of the players in the market”. Their organizational structure is built to deal with

SC complexity as “{Plant H} is controlled by entrepreneurs. The employees are free to do what

they want, they aren’t controlled that much. They are treated like entrepreneurs”. Indicating, also

by their hierarchical level of 3-4, that employees are free to deal with complexity drivers. All other plants in this category score medium on SC complexity. These plants have in common that they have a limited amount of hierarchical levels, except for plant B that, likely through its Chinese culture, has seven hierarchical levels. Its sales manager states that “The organizational

structure is planned in advance by the company. Each department has its own duty. In addition, each department has its own KPIs, restraining it to its own performance” (Sales manager, Plant

(20)

20 its medium size and efficient work order still refrains it from experiencing a high level of SC complexity.

The final category comprises the last three plants, which are all relatively large in comparison to the before mentioned organization. Two out of the three plants have a large number of employees, they are all producing over a 100.000 tons of product a year, have at least five plants or more and a relatively high number of processing steps. As mentioned previously in the findings section, these plants (A, D, G) all process large quantities of natural products. However, the organization show a difference in terms of SC complexity within their plants, where D scores low and both A and G score relatively high. This difference is not found in the organizational structure as these are relatively equal, but in the amount of different products they create and the way in which they cooperate with their suppliers. For instance, Plant D has indicated to be “a

production location” as has the logistical manager of plant G “we are purely a production plant”

and “the only thing we make is deep frozen fries, although in different cutting sizes … that is not

difficult whatsoever”. This is different from plant A that says to “have around 1400 end products” (Planning and distribution manager). However, this does not explain why plant D

experiences less complexity than Plant A and G. This is found in the way in which it handles its raw materials, and in particular its quality control as was explained under ‘FPI characteristics’ at the beginning of this chapter.

4.3.2 Formalization vs. SC complexity

When it comes to the next construct, all organizations were found to experience a medium to high level of formalization. This is mostly due to organizations indicating to have strict rules and regulations within their production facilities. “We can’t just walk into the factory wearing our

normal clothing. Safety shoes, hair nets, washing your hands, all the hygiene that is necessary.”

(Logistical Manager, Plant G). The results are depicted in table 8.

Table 8 – Formalization within the sample

The third finding is quite interesting. Where every organization indicated that they have to deal with strict rules and regulations in their production process, ranging from “everything has to deal

with quality and food safety of the end products” (Purchasing manager, Plant H), to “The requirements are very high and are almost as high as the pharmaceutical industry” (Supply

chain manager, Plant D) and the supply network planner of Plant E indicated that they “had to

place a new fence around the premises with 24/7 security in order to be able to do business with Russia after the current trade-embargo”. However, there is a big difference in culture in relation

to formalities, rules & regulations and the level of innovation. This pattern can be extended to the level of

interaction # regulations # rules

(21)

21 previous construct, where smaller organizations are rather more informal than big corporations. One manager from plant G indicated just that by saying: “But once you start in a smaller

company like {Plant G}, where you get the room to do things. Then that is way cooler than working in a big corporation that moves slowly to the left or slowly the right where you are no more than a small radar blip on a big spectrum and where you have no influence”. In addition,

the supply chain manager of Plant D, a big corporation in infant nutrition, indicates that when

“you compare it with other companies in the food industry, we are very formal”. However, this

does not hold for the remaining two large organizations that indicate to be informal, whilst limited in their decision making “we need to follow a lot of procedural (sequences) steps before

something actually changes” (Manager sourcing & contracting, Plant A). It is found that the

difference between formal and informal organizations has no influence on the experienced SC complexity.

4.4 Vertical integration vs. SC complexity

Within the sample of 10 organizations, only one plant has indicated to be fully integrated “We as

dairy company are very unique because we dominate the whole supply chain … it starts with the cow and ends with what is provided to the baby by its mother” (Supply chain manager, Plant D).

In addition, there is another manager stating that “since we have our own retail stores,

downstream is partly integrated” (Quality manager, Plant J). However, overall the organizations

from the sample are limited to a single step “The crumbs is a step forward, in the past we thought

we had to go way more towards the customer … we are now convinced that it would be better to take a step back” (Purchasing manager, Plant F). The statement made by the purchasing manager

from Plant F seems to be widely carried among the different organizations, making up our fourth finding. For instance, the manager planning & logistics from Plant H indicates that “There is a

special relationship with the upstream supply chain, however we don’t own those tiers”. This is

mostly the result of the customers, both businesses and consumers, wanting to know the origin of the product and to ensure continuous quality in the end-product. However, from the current sample no indication is given that vertical integration influences SC complexity as all plants that experience high levels of SC complexity are very little to partly integrate. In addition, the highest integrated organization, plant D, scored very low on SC complexity.

Figure 2 – level of vertical integration sample organizations.

Plant A Plant B Plant C Plant D Plant E Plant F Plant G Plant H Plant I Plant J Fully integrated No integration Little integration Partly

(22)

22

Discussion

The data analysis has enabled for insights related to all facets of the paper’s theoretical framework. The findings from the previous section will be discussed in the same sequence as in which they were reported. Therefore, starting with the FPI characteristics, followed by SC complexity and its relationship to FPI, the different organizational structure constructs and their link to FPI and the relationship to SC complexity. These findings will be compared or related to literature and this paper’s conceptual model and accompanying research question.

5.1 FPI Characteristics

The assessment of the interviews has led to an understanding of the influence of FPI characteristics on the sample. The findings rather confirm previous research by Van Donk (2001) in terms of perishability and changes in customer wishes being the main characteristics that influence the industry. In addition, organizations are seen as having to deal with an increasing amount of rules and regulations [Garcia Martinez et al., 2006; Mahalik & Nambiar, 2010; Sarpong, 2014; Trienekens & Zuurbier, 2008] that are equally becoming stricter further putting a strain on SC activities. Furthermore, it was found that variability in supply (yield/harvest) and quality is often mentioned within organizations that deal with a natural product (Van Donk, 2001). However, there seem to be differences between these organizations, whereas some organizations are found to better deal with this variability than others. This can be related to collaboration with suppliers (Manuj & Sahin, 2011) through information sharing as indicated by Plant G, and developing technologies together with suppliers like plant D.

5.2 SC Complexity within FPI

(23)

23

5.3 Organizational structure - configuration

Zhou (2002) stated that the number of hierarchical levels could have either a positive or negative influence on decision making process of an organization. It was found that the way in which the organizations are configured has a limited influence on SC complexity. More so, it can be noted that the organizations have adapted their organizational structure to cope with the FPI characteristics as identified previously. This show in the relative low level of hierarchical levels across the sample, these shorts lines and responsibilities across the departments make for quick decision making when it comes to dealing with FPI characteristics. One or the organizations stood out with its seven hierarchical levels, but still reported to have a medium level of SC complexity. However, this finding is not sector specific as others researchers [Lam & Chin, 2005; Aissaoui, Haouari, & Hassini, 2007] have reported the negative influence of large hierarchies on a firm’s performance. In addition, based on the difference in hierarchical levels between the organizations in the sample, one is unable to form a conclusive note. Therefore, the configuration cannot be regarded to as having an influence on SC complexity within the FPI.

5.4 Organizational structure – formalization

Upon assessing a sample of organizations active in the FPI, especially those dealing with infant nutrition, one would expect very tight regimes concerning the rules and regulations in both decision making and procedures. However, this only partly holds for our sample as the production facilities where unanimously very strict, but differences were found in how things where run outside the factories. This difference should have had a similar effect on the level of SC complexity, where formal organizations are less flexible due to limited communication and interaction between organizational members (Pertusa-Ortega et al., 2010). Consequently, it is expected that formal organizations show an increase in SC complexity. However, the sample shows no indication that the formalization of the organizations has an influence on the level of SC complexity. This can be explained through the fact that, however informal the organization, all decisions, innovations and procedures have to be subject to very strict rules and guidelines concerning food, health and safety regulations [Garcia Martinez et al., 2006; Mahalik & Nambiar, 2010; Sarpong, 2014; Trienekens & Zuurbier, 2008]. This was observed through many of the organizations where managers indicated to be very free in their job, but where new ideas where subject to years of testing and legal work. Hence, it can be concluded that FPI characteristics in regard to food and safety regulations negatively influence the organizational structure in terms of speed and flexibility.

5.5 Organizational structure - vertical integration

(24)

24

Conclusion

This research was set out to provide an answer to the research questions: How does an

organization’s structure influence the supply chain complexity within the food processing industry?

The literature review has shown that consensus on how to measure SC complexity had not yet been reached and therefore a recent model by De Leeuw et al. (2013) was chosen to identify SC complexity drivers. In addition, literature shows a gap in regard to identifying the actual influence and manageability of said drivers on SC complexity (Manuj & Sahin, 2011). It is for this reason that the researcher, through an aggregated model based on earlier findings by James & Jones (1976), have identified the sample’s organizational structures. Accordingly, the FPI characteristics where identified through publications by Van Donk (2001), Van Wezel et al. (2006), Kilic et al. (2013) and (Trienekens & Zuurbier, 2008). These constructs have resulted in a conceptual model to test the relationship between the organizational structure and SC complexity based on the influence of FPI characteristics.

In order to fill the gap, an exploratory research was done in the form of a multiple case study amongst 10 organizations within the FPI. These organizations show both similarities and differences in terms of their organizational structure and the product they produce. In order to obtain a complete overview of the constructs, the researchers aimed to have two interviews per organization with two different managers.

Theoretical implications

From a theoretical point of view, by combining the insights of several authors with the results of a multiple case study, this research was the first to investigate the relationship between the organizational structure as an individual complexity driver and its influence on the overall SC complexity within the FPI. Throughout the study a large variety of FPI characteristics have been identified and confirmed to have an influence on both the organizational structure and SC complexity. However, no such relation has been found that shows a significant influence between the organizational structure and the level of SC complexity. For one, the configuration of the organizations, although different, have shown no relationship with the level of SC complexity as experienced by the organizations under study. In addition, although the strict rules and regulations in both the decision making and production within the FPI make the organizations increasingly formalized, it does not influence the level of SC complexity. As a final result regarding the organizational structure, due to the low level of vertical integration within the sample, no indication is given that vertical integration influences the experienced level of SC complexity. It was also found that the results in the study might be skewed due to the lack of a sector specific measurement tool. It has therefore identified the need to create such a tool that can be used to assess specific sector on their SC complexity and can help in finding fitting manageable strategies to mitigate SC complexity.

Managerial implications

(25)

25

Limitations and proposed future research

The research has known a couple of limitations. For one, some organizations were unable to supply two managers for our interviews. Therefore, some interviews revolve around one person’s knowledge and opinion, making it difficult to verify the quality of the various constructs. In addition, interviewing one respondent might increase the possibility of bias. A second limitation relates to the fact that the interview protocol was comprised of five different researches, all steering on their own topic when conducting an interview. This has resulted in a less detailed overall overview of the organizations.

(26)

26

Bibliography

Acemoglu, D., Griffith, R., Aghion, P., & Zilibotti, F. (2010). Vertical Integration and Technology: Theory and Evidence. Journal of the European Economic Association,

8(September), 989–1033.

Aissaoui, N., Haouari, M., & Hassini, E. (2007). Supplier selection and order lot sizing modeling: A review. Computers and Operations Research, 34, 3516–3540.

Anderson, C., & Brown, C. E. (2010). The functions and dysfunctions of hierarchy. Research

in Organizational Behavior, 30(C), 55–89.

Bakker, F., & Donk, D. P. Van. (2012). Identifying barriers to internal supply chain integration using Systems Thinking, 1–10.

Bozarth, C. C., Warsing, D. P., Flynn, B. B., & Flynn, E. J. (2009). The impact of supply chain complexity on manufacturing plant performance. Journal of Operations

Management, 27, 78–93.

Chen, C. J., & Huang, J. W. (2007). How organizational climate and structure affect knowledge management-The social interaction perspective. International Journal of

Information Management, 27, 104–118.

Choi, T. Y., Dooley, K. J., & Rungtusanatham, M. (2001). Supply networks and complex adaptive systems: Control versus emergence. Journal of Operations Management, 19, 351–366.

De Leeuw, S., Grotenhuis, R., & Goor, A. R. Van. (2013). Assessing complexity of supply chains: evidence from wholesalers. International Journal of Operations & Production

Management, 33, 960–980.

Eisenhardt, K. M., & Graebner, M. E. (2007). THEORY BUILDING FROM CASES : OPPORTUNITIES AND CHALLENGES. Academy of Management Journal, 50(1), 25–32.

Frizelle, G., & Woodcock, E. (1995). Measuring complexity as an aid to developing operational strategy. International Journal of Operations & Production Management,

15, 26–39.

Gallo, P. J., & Christensen, L. J. (2011). Firm Size Matters: An Empirical Investigation of Organizational Size and Ownership on Sustainability-Related Behaviors. Business &

Society, 50(2), 315–349.

Garcia Martinez, M., Poole, N., Skinner, C., Illés, C., & Lehota, J. (2006). Food safety performance in European Union accession countries: benchmarking the fresh produce import sector in Hungary, 25(2), 231–250.

Georgiadis, P., Vlachos, D., & Iakovou, E. (2005). A system dynamics modeling framework for the strategic supply chain management of food chains. Journal of Food Engineering,

(27)

27

Gerschberger, M., Engelhardt-nowitzki, C., Kummer, S., & Staberhofer, F. (2012). A model to determine complexity in supply networks. Journal of Manufacturing Technology

Management, 23(8), 1015 – 1037.

Hansen, G. S., & Wernerfelt, B. (1989). Determinants Of Firm Performance: The Relative Importance O. Strategic Management Journal, 10(5), 399–411.

Jacobs, M. a. (2013). Complexity: Toward an empirical measure. Technovation, 33(4-5), 111–118.

James, L. R., & Jones, A. P. (1976). Organizational structure: a review of structural dimensions and their conceptual relationships with individual attitudes and behavior.

Organizational Behavior and Human Performance, 16, 74–113.

Karantininis, K., Sauer, J., & Furtan, W. H. (2010). Innovation and integration in the agri-food industry. Food Policy, 35(2), 112–120.

Karlsson, C. (2009). Research Operations Management. Routledge, New York.

Kern, A. (2006). Exploring the Relation Between Creativity and Rules: The Case of the Performing Arts. International Studies of Management and Organization, 36(1), 63–80.

Kilic, O. a., Akkerman, R., van Donk, D. P., & Grunow, M. (2011). Intermediate product selection and blending in the food processing industry. International Journal of

Production Research, 51(1), 1–17.

Lacity, M. C., & Janson, M. a. (1994). Understanding Qualitative Data: A Framework of Text Analysis Methods. Journal of Management Information Systems, 11(2), 137–155.

Lam, P. K., & Chin, K. S. (2005). Identifying and prioritizing critical success factors for conflict management in collaborative new product development. Industrial Marketing

Management, 34, 761–772.

Mahalik, N. P., & Nambiar, A. N. (2010). Trends in food packaging and manufacturing systems and technology. Trends in Food Science & Technology, 21(3), 117–128.

Maloni, M. J., & Brown, M. E. (2006). Corporate social responsibility in the supply chain: An application in the food industry. Journal of Business Ethics, 68, 35–52.

Manuj, I., & Sahin, F. (2011). A model of supply chain and supply chain decision-making complexity. International Journal of Physical Distribution & Logistics Management (Vol. 41).

Matt, D. T. (2007). Reducing the structural complexity of growing organizational systems by means of axiomatic designed networks of core competence cells. Journal of

Manufacturing Systems, 26(2007), 178–187.

Meredith, J. (2002). Introduction to the special issue : case study and field research. Journal

(28)

28

Michael, S. C. (2000). The effect of organizational form on quality: the case of franchising.

Journal of Economic Behavior & Organization, 43(3), 295–318.

Nayak, N. V., & Taylor, J. E. (2009). Offshore Outsourcing in Global Design Networks.

Journal of Management in Engineering, 25(October), 177–184.

Perona, M., & Miragliotta, G. (2004). Complexity management and supply chain performance assessment. A field study and a conceptual framework. International

Journal of Production Economics, 90, 103–115.

Pertusa-Ortega, E. M., Zaragoza-Sáez, P., & Claver-Cortés, E. (2010). Can formalization, complexity, and centralization influence knowledge performance? Journal of Business

Research, 63(3), 310–320.

Sarpong, S. (2014). Traceability and supply chain complexity: confronting the issues and concerns. European Business Review, 26, 271–284.

Sivadasan, S., Efstathiou, J., Calinescu, a., & Huatuco, L. H. (2006). Advances on measuring the operational complexity of supplier-customer systems. European Journal of

Operational Research, 171, 208–226.

Trienekens, J., & Zuurbier, P. (2008). Quality and safety standards in the food industry, developments and challenges. International Journal of Production Economics, 113, 107–122.

Vachon, S., & Klassen, R. D. (2002). An exploratory investigation of the effects of supply chain complexity on delivery performance. IEEE Transactions on Engineering

Management, 49(3), 218–230.

Van Donk, D. P. (2001). Make to stock or make to order: The decoupling point in the food processing industries. International Journal of Production Economics, 69, 297–306.

Van Wezel, W., Van Donk, D. P., & Gaalman, G. (2006). The planning flexibility bottleneck in food processing industries. Journal of Operations Management, 24, 287–300.

Yin, R. K. (2009). Case study research: design and methods. Thousand Oaks, CA: Sage

Publications, Inc.

Zhou, D. Z. D. (2002). An empirical study of the role of postponement application in reducing supply chain complexity. IEEE International Engineering Management

ORGANIZATION OF THE FOOD PROCESSING INDUSTRY AND ITS IMPACT ON SUPPLY CHAIN COMPLEXITY: A MULTIPLE CASE STUDY