INOM
EXAMENSARBETE INDUSTRIELL EKONOMI,
AVANCERAD NIVÅ, 30 HP ,
STOCKHOLM SVERIGE 2020
Addressing shortcomings in
goods distribution
A case study of a collaborative distribution
network
AXEL THÅLIN
VICTOR GJERS
KTH
Addressing shortcomings in goods
distribution
A case study of a collaborative distribution network
by
Victor Gjers
Axel Thålin
Master of Science Thesis TRITA-ITM-EX 2020:246 KTH Industrial Engineering and Management
Hantering av tillkortakommanden inom
varudistribution
En fallstudie av ett distributionssamarbete
av
Victor Gjers
Axel Thålin
Examensarbete TRITA-ITM-EX 2020:246 KTH Industriell teknik och management
Master of Science Thesis TRITA-ITM-EX 2020:246
Addressing shortcomings in goods distribution
A case study of a collaborative distribution network
Victor Gjers Axel Thålin Approved 2020-06-03 Examiner Jannis Angelis Supervisor Luca Urciuoli Commissioner
Pernod Ricard Sweden AB
Contact person
Mikael Andersson
Abstract
In the increasingly frequent distribution of the Fast-Moving Consumer Goods (FMCG) industry, there is an evident need of improving supply chain operations to ensure sustainability. As part of the supply chain, the distribution of goods often accounts for a major part of costs and resources. Therefore, focus needs to be directed toward ensuring a continuously functional distribution network. Performance measures of high asset utilization and high customer service level are often contradictory to some extent, making the improvement work even more complex. One way of increasing the performance of distribution operations is to take part in an interorganizational collaboration.
The purpose of this master thesis is to investigate how the distribution operations of a FMCG company connected to a collaborative distribution network could be improved. Through a single case study, the researchers identify shortcomings in the distribution operations of a cross-docking network with a focus on a specific customer segment flow. By using concepts of demand management, supply chain collaboration, and customer satisfaction, the findings are analysed to highlight shortcomings in the distribution.
Researchers discuss how distribution performance consequences could be related to certain managerial actions by presenting three different scenarios. The scenarios show a distinct effect on the number of shipments conducted towards cross-dock. As the number of shipments are reduced, economic benefits could be realized. However, these benefits must be carefully compared with the potential reduction in the level of customer service. The conclusion of this study suggests that managerial action is necessary to increase the economic sustainability of the distribution.
Key-words
Examensarbete TRITA-ITM-EX 2020:246
Hantering av tillkortakommanden inom varudistribution
En fallstudie av ett distributionssammarbete
Victor Gjers Axel Thålin Godkänt 2020-06-03 Examinator Jannis Angelis Handledare Luca Urciuoli Uppdragsgivare
Pernod Ricard Sweden AB
Kontaktperson
Mikael Andersson
Sammanfattning
Med den allt mer ökande distributionsfrekvensen inom dagligvaruhandeln finns det ett tydligt behov av att av att förbättra värdekedjeoperationerna för att säkerhetsställa hållbarhet. Distribution av det fysiska varuflödet är en viktig del av värdekedjan och står för en stor del av de totala kostnaderna och resursanvändningen. Det är därför väsentligt att fokusera på att säkerhetsställa ett funktionellt distributionsnätverk som presterar kontinuerligt. Hög nyttjandegrad samt hög kundnöjdhet är två prestationsgrundade objekt som är svåra att uppnå parallellt vilket försvårar ett förbättringsarbete. En metod för förbättring är att flera dagligvaruhandelsbolag går in i ett distributionssamarbete.
Syftet med denna masteruppsats är att undersöka hur distributionsverksamheten i ett dagligvaruhandelsbolag som är kopplat till ett kollaborationsdistributionsnätverk kan förbättras. Genom en fallstudie identifierades flera tillkortakommanden i distributionsverksamheten i en del av varuflödet kopplat till en kundgrupp. Resultaten analyserades med hjälp av koncept från kundnöjdhet, kollaborationsvärdekedjor samt hantering av efterfrågan för att belysa tillkortakommanden inom distributionsverksamheten.
Genom tre olika scenarion diskuteras olika konsekvenser på distributionsverksamheten som kan kopplas till olika åtgärder. Scenarierna visar att antalet genomförda leveranser till mellanlager har en stor inverkan på hur väl distributionsverksamheten presterar ekonomiskt. Studien föreslår sammanfattningsvis att olika beslutsåtgärder är nödvändiga för att öka den ekonomiska hållbarheten i distributionen.
Nyckelord
Table of Content
1 Introduction ... 1 1.1 Background ... 1 1.2 Problem formulation ... 2 1.3 Case problem ... 2 1.4 Purpose ... 2 1.5 Research question ... 3 1.6 Delimitations ... 3 2 Theoretical Framework ... 4 2.1 Distribution networks ... 42.2 Collaborative supply chain... 7
2.3 Customer satisfaction... 8 2.4 Demand Management ... 11 3 Method ... 15 3.1 Choice of Methodology ... 15 3.2 Research Design ... 15 3.3 Quality of research ... 21 3.4 Ethical considerations ... 22 4 Case Background ... 23
5 Results & Analysis ... 25
5.1 Distribution Network ... 25
5.2 Customer Service Level ... 27
5.3 Collaboration and synchronization ... 28
5.4 Past demand ... 29
5.5 Scenarios ... 35
6 Discussion ... 41
6.1 Distribution network ... 41
6.2 Customer Service Level ... 41
6.3 Collaboration and Synchronization... 42
6.4 Past Demand... 43
6.5 Scenarios ... 44
7 Conclusion ... 47
7.1 Fulfilling the purpose of the research ... 47
7.2 Contribution ... 48
7.3 Reflections on sustainability ... 49
7.4 Limitations of the study and further research ... 49
8 References ... 50
APPENDIX 1 – Daily delivered volume to Trollhättan 2019... 1
List of Tables
Table 1 - Cross-dock problem classes and descriptions (Buijs, et al., 2014) ... 6
Table 2 - Observations conducted in the context study ... 16
Table 3 - List of conducted interviews ... 18
Table 4 – Pivot matrix ... 19
Table 5 - Company responsibilities in the distribution network ... 23
Table 6 –Summary of results from a customer survey conducted in 2017 (Internal Data) ... 24
Table 7 – Summary effects from the different scenarios ... 39
List of Figures
Figure 1 - Single cross-dock network configurations (Buijs, et al., 2014) ... 5
Figure 2 – Framework on customer satisfaction, adapted from Chavez et al. (2016) ... 8
Figure 3 - The Demand Management Process (Croxton, et al., 2002) ... 12
Figure 4 - Synchronizing Process adopted from (Croxton, et al., 2002) ... 14
Figure 5 – Research Process ... 15
Figure 6 - Distribution network configuration ... 25
Figure 7 - Geographical locations of cross-docks and central warehouse ... 26
Figure 8 – Percentage of shipments above and below 100L, all cross-docks... 30
Figure 9 – Graph showing the relation between delivered volume, cost and earnings. ... 30
Figure 10 – Shipments to cross-docks in major cities ... 31
Figure 11 – Shipments to cross-docks in medium and minor cities ... 31
Figure 12 – Total delivered volume per day, 2019 ... 31
Figure 13 – Total delivered volume per day 2019, major cities... 32
Figure 14 – Total delivered volume per day 2019, medium and minor cities ... 32
Figure 15 – Percentage of shipments above and below 100L, Trollhättan ... 33
Figure 16 - Weekly volume at Trollhättan 2019, starting at week 10. ... 33
Figure 17 – Total delivered volume per weekday to Trollhättan 2019 ... 33
Figure 18 – Shipment volume performance, Trollhättan, whole year ... 34
Figure 19 – Summer, week 27-32 ... 34
Figure 20 – Weekly performance of shipments. ... 35
Figure 21- Scenario 1 percentage of shipments above and below 100L, Trollhättan ... 36
Figure 22 – Performance, Trollhättan, Scenario 1 ... 36
Figure 23 - Scenario 2 percentage of shipments above and below 100L, Trollhättan ... 37
Figure 24 – Performance, Trollhättan, Scenario 2 ... 37
Figure 25 - Scenario 3 percentage of shipments above and below 100L, Trollhättan ... 38
Foreword
This master thesis was conducted during January to June 2020 in cooperation with Pernod Ricard Sweden AB, at the Department of Industrial Engineering and Management at the Royal Institute of Technology, Stockholm.
Acknowledgements
Firstly, we would like to thank our supervisor Luca Urciuoli, who throughout the work with the thesis provided guidance, support, and helpful insights. By sharing his knowledge within the field of research he helped us to increase the quality of our research. Further, we want to express our gratitude towards the case company representatives and other individuals who showed great will of cooperation throughout the study. We want to direct a special thanks to Mikael Andersson at Pernod Ricard Sweden AB, for being available throughout the study, answering questions, and helping us to set up interviews and observations. Finally, we want to thank our examiner and seminar leader Jannis Angelis together with our peers, who provided valuable feedback that has improved our work, and the outcome of it.
Stockholm, June 2020
1 Introduction
This chapter presents a background to the area of research followed by the problem formulation, purpose, and research questions of the study. Lastly it presents the delimitations and purposed contributions of the study.
1.1 Background
The global beverage industry is increasing in volume and the global beverage market is projected to be valued at $ 1.9 trillion by 2021 (Jones & Comfort, 2019). Within the beverage industry alcoholic beverages accounts for 64% of the total value (Holmes & Anderson, 2017). Food, beverages, and household products are usually grouped together as different segments of the Fast-Moving Consumer Goods (FMCG) industry. An industry that is characterised by goods that are sold frequently, and produced at high volume and low cost (Aljunaidi & Ankrah, 2014).
In the competitive landscape of any market, the supply chain plays a vital role, as “one of the
most critical disciplines in the nowadays global business world” (Lapinskaitė & Kuckailytė,
2014). In the area of FMCG, research points out inefficiencies in the distribution of goods to the final customer (Ranieri, et al., 2018; Macioszek, 2017). Often, the last part of the logistic operations account for an integral part of the total cost, highlighting the importance of efficient logistic operations in the latter stages of the supply chain.
The high frequency of FMCG requires the actors within the industry to distribute their goods on a daily basis to their customers. This frequent distribution is made possible by operating in networks to distribute the goods efficiently from few locations to several customers. The transport fleet usually consists of trucks due to the geographical spread of customers (Kellner, et al., 2013). One strategy used within the FMCG industry is cross-docking networks. In a distribution network, cross-docks are used as consolidation points, to group shipments from several retailers into one truckload which then distribute the goods to several end customers close to the cross-dock (Buijs, et al., 2014; Kellner, et al., 2013). The distribution from the cross-dock to the end customer is often called the last mile-delivery or last mile-distribution, where a truck drives a predetermined route dropping of goods at several customers.
Collaboration initiatives differs within the different segments of FMCG. Within the beverage segment, actors may collaborate through combining their product portfolios e.g. a wine & spirits and a beer & soft drink company to provide their customers a wide selection of products to receive in a single shipment from a cross-dock. Further, it occurs that competitors within the wine & spirits segment share warehouses and distribution towards cross-docks and customers. However, a lack of collaborative planning within the collaborative supply chain have a negative impact on the overall supply chain performance (Attaran & Attaran, 2007). Different participants have separate interests and may not recognize the same problems as another member of the collaborative supply chain is experiencing.
1.2 Problem formulation
The supply chain, and more specifically, the distribution networks of FMCG companies are pressured to provide a high service level (Nozari, et al., 2019), in terms availability, i.e., number of delivery days, order cut-off times and lead times. These measures are highly relevant to stay competitive in terms of providing customer value. However, there might be a need to adjust these measures to reach a certain level of distribution utilization in order to retain economic and environmental sustainability. Chen & Chankov (2017) emphasise that there is a conflict in the goals of last-mile delivery, as high asset utilisation and high service level towards the customer are contradictory to some extent (Chen & Chankov, 2017). For firms operating in a collaborative distribution network, assessing this problem becomes increasingly complex as several actors are involved, with varying organizational objectives and operational conditions (Zhang & Cao, 2018).
1.3 Case problem
The case company is one of the world leading wine and spirits retailers which operates in Sweden through a subsidiary. In Sweden, their business is divided into two segments, on-trade, and off-trade. Off-trade includes all the retail where their product can be bought in a physical store and on-trade is the retail towards restaurants.
Currently, the case company is experiencing issues in their distribution towards the cross-docks serving their on-trade customers. The high service level forces the case company to distribute goods to cross-docks on a daily basis where shipment volumes can vary from below five litres, occupying a pallet spot on a truck, to several pallets of more than 300 litres in a single shipment. This is causing several of the shipments to be unsustainable from economic and environmental perspectives.
1.4 Purpose
1.5 Research question
To fulfil the purpose of the study and help guide the research, two research questions were formulated. Research question 1 (RQ1) was formulated to assist in identifying the shortcomings in the distribution towards cross-docks. Research question 2 (RQ2) aims to guide the research to provide deeper insights in how the shortcomings can be addressed and what effects that might be experienced from possible changes in the distribution.
RQ1:
What shortcomings are there in the current distribution towards cross-docks?
RQ2:
How could a FMCG company, operating in a collaborative distribution network, increase its on-trade distribution performance while maintaining a satisfactory service level towards customers?
1.6 Delimitations
The study has followed the delimitations stated below.
• The study has been delimited towards the alcoholic beverage branch within the FMCG industry and was based on a single case study at a large wine & spirit producer.
• The main focus of the analysis has been the distribution from the central warehouse toward cross-docks and the shortcomings related to it. However, the researchers have also considered factors related to the distribution network as a whole.
• The study only covers on-trade customers connected to the collaborative ordering system, i.e., restaurants.
• The quantitative data collected and analysed in this study has been delimited to the distributed volumes from the central warehouse to the cross-docks. The data regards volumes from the year of 2019.
2 Theoretical Framework
This chapter introduces a literature review of logistic concepts and strategies relevant to the study including distribution networks and supply chain collaboration. Further, it describes the concept of customer satisfaction, and lastly, a section describing demand management is presented. The concepts and subjects presented in the literature review form the theoretical framework of this study.
2.1 Distribution networks
A distribution network can be described as the nervous system of a supply chain. It determines the flow of products through storage facilities and transportation alternatives. There are several options to consider when designing a distribution network. Depending on e.g. product variety, volume, distance and customer needs, different decisions regarding strategy and layout of the distribution network must be considered.
2.1.1 Cross-docking
Cross-docking is a strategy used in the distribution of goods to aggregate and consolidate less-than-truckloads of goods into full truckloads (Buijs, et al., 2014). In contrast to a conventional warehouse or distribution centre, a cross-dock does not offer any long-time storage of goods. Instead, it acts as a hub, where shipments from different locations, warehouses and distribution centres, are unloaded, reorganized and loaded onto trucks to distribute goods with a geographical proximity of destinations (Boysen & Fliedner, 2010). Cross-docking is appropriate in the industry of fast-moving goods such as perishables and, where the demand is relatively constant or has a low variance (Hasani Goodarzi & Zegordi, 2017). However, there are examples of cross-docks being used in other industries as well, improving lead times, stock levels and transportation costs (Serrano, et al., 2017).
A central part of the cross-docking strategy is the network design. In a simplified cross-dock network, the cross-dock can be represented as a node connecting a number of suppliers to a number of destinations. Figure 1, adapted from the work of Buijs et al. (2014), visualises different single cross-dock configurations. These three configurations are each suitable under different circumstances, depending on the type of products that are transported. The
many-to-few network configuration is often encountered in contexts of manufacturing, where a many-to-few
manufacturing plants receive components and material from several suppliers. In a few-to-many
network configuration goods from a few distribution centres are consolidated and distributed to
Figure 1 - Single cross-dock network configurations (Buijs, et al., 2014)
A more advanced configuration of a cross-dock network is the hub-and-spoke system including multiple cross docks. This network configuration is typical in the presence of less-than- truckload shipments with varying volumes (Stephan & Boysen, 2011). The network usually consists of a central hub represented by a central warehouse, distribution centre or cross-dock. From the central hub shipments are distributed to different cross-dock locations, either as a single point-to-point route or as a succession of deliveries. At each cross-dock the goods of each shipment are consolidated to an outbound truck and delivered to a final destination. Cross-docking can contribute to an increase of vehicle utilization, reducing the transportation costs (Hasani Goodarzi & Zegordi, 2017).
Table 1 - Cross-dock problem classes and descriptions (Buijs, et al., 2014) Problem classes Description
Network design Decisions regarding the cross-docking network’s physical infrastructure.
Network planning Decisions regarding logistic resources, in terms of allocation and utilization, on a network level such as the routing of freight.
Network scheduling Decisions regarding dispatchment and alignment of shipments.
Cross-dock design Decisions regarding exterior and interior design and configuration of the cross-dock.
Cross-dock planning Decisions regarding medium-term operations such as determining the appropriate workforce and minimizing material handling.
Cross-dock scheduling Decisions regarding timing and sequencing at the cross dock.
2.1.2 Last mile logistics
Last-mile logistics is a term used in logistics and supply chains to describe the transportation of goods from a hub to the final customer. It refers to the last stretch delivery of a parcel in a business-to-consumer arrangement (Lim, et al., 2015). However, it might also include arrangements where the end point of the logistic operation is not a consumer but a collection point such as a pick-up location, dealer, or restaurant. The last mile logistics of a supply chain is usually an inefficient stage and can account for up to 28% of the total cost of delivery (Ranieri, et al., 2018). One reason for this inefficiency is that deliveries usually consist of small quantities at each delivery point which prevents leveraging the economy of scale (Deutsch & Golany, 2018).
2.2 Collaborative supply chain
Matching the supply and demand can be a difficult task, especially for actors dealing with short lifecycle products (Simatupang & Sridharan, 2002). However, this problem could be addressed through a supply chain collaboration. A collaborative supply chain offers the opportunity to achieve more effective operations, but also poses risks and challenges to the collaborating actors. Cao and Zhang (2011) synthesize from previous literature, seven value adding dimensions in the collaborative supply chain: information sharing, goal congruence, decision synchronization, incentive alignment, resources sharing, collaborative communication, and joint knowledge creation (Cao & Zhang, 2011).
Information sharing is by many researchers described as a vital pillar in a collaborative supply chain. This dimension is enabled by information- readiness, accuracy, and security which all play a role in the overall outcome of the collaboration (Panahifar, et al., 2018). Information sharing is the dimension which all other dimensions stem from. It should be noted that trust between the parties of the collaboration is an influential factor (Fawcett, et al., 2015). Creating a well-functioning IT infrastructure to connect the collaborating partners is important since it enhances the information sharing (Wu, et al., 2014). Research highlights that such inter organizational system is a key stone in a supply chain collaboration (Zhang & Cao, 2018). The flows of material, information and finance are all facilitated by a well-functioning IT infrastructure. Research has also pointed out that in supply chain collaboration, IT enhances both incremental and radical product innovation (Jimenez-Jimenez, et al., 2019).
The organizational culture of individual firms in a collaborative supply chain can act either as an enabler or a disabler (Cao & Zhang, 2013). It is not uncommon that organizations have different short-term interests which can cause conflict and obstruct the collaborative efforts (Holweg, et al., 2005). Norms and beliefs of organizations in a collaboration must be aligned to the extent that it promotes the collaboration. This also emphasises the importance of common goals in the collaboration. The collaborative culture has been shown to have a significant impact on supply chain collaboration. Drivers that enable the collaborative culture include collectivism, long term orientation, power symmetry and uncertainty avoidance (Zhang & Cao, 2018).
2.3 Customer satisfaction
Customer satisfaction is essential to understand within a supply chain. If the expected service demand from customer is met it is more likely for the company to obtain future sales from the same customer (Abu-Salim, et al., 2017). To successfully meet the expected service demands, a company needs to understand the customer needs to define the adequate service levels that they should maintain (Chopra & Meindl, 2016). Quality, cost, delivery and flexibility are four areas of service level that can affect the customer satisfaction in different ways (Chavez, et al., 2016). Further Chavez, et al. (2016) suggest that customer involvement in making the supply chain more environmentally friendly has a positive effect on the four mentioned areas of service level.
Figure 2 show a theoretical framework on customer satisfaction that has been adapted from Chavez et al. (2016) and the elements within are further described below.
Figure 2 – Framework on customer satisfaction, adapted from Chavez et al. (2016) 2.3.1 Service quality
Service quality is commonly conceptualized as the distance between the customers’ expectations and the evaluation of the perceived service. This can be measured by evaluating five key categories: responsiveness, assurance, reliability, tangibles and empathy (Seth, et al., 2006; Cronin & Taylor, 1992; Parasuraman, et al., 1988).
The five key categories were identified by Parasuraman et al. (1985; 1988) who describes the different factors that affects each category:
• Assurance is dependent on the knowledge, competence, and courtesy of the employees of the company. For a customer, it is important that their contact persons from the company have a high level of skill and knowledge and that the personnel working operationally has the adequate competence to provide the service.
• Reliability is linked to the company’s ability to honour its promises towards the customer. It involves the company’s ability to deliver the promised service correctly, within the agreed time.
• Tangibles are the physical evidence of the service provided. It includes everything from the appearance of the personnel delivering the service to the representation of the service, such as how the delivered goods are packed.
• Empathy is linked to what extent the company is willing to provide individualised caring solutions for a customer. Further it is also affected by the company’s ability to consistently recognize the customer and learn the customer specific requirements.
Nonetheless, Seth et al. (2006) empathises that with the global marketplace, service quality becomes more difficult to manage. Many companies do not own logistic assets in their supply chain, as it is being outsources to third-party logistics (3PL). The use of 3PLs can create gaps in what quality the company want to deliver and what quality that is actually delivered. The role of the 3PL in the supply chain is affected by to what extent it is needed, the perception of the customer, and what relationship the 3PL has with the customer.
2.3.2 Cost of service
The overall objective of most companies is to recover their total costs and make a profit when they sell their goods. The profit in question is dependent on what the customer is prepared to pay for the total benefits from the goods and service. The value perceived for the customer is usually conceptualized as the difference between the total cost and total benefit (Abu-Salim, et al., 2017). Further, Abu-Salmin et al. (2017) suggests that the total cost can be evaluated as the economic cost (the price paid), time cost (the time it took to decide and receive the service), human energy cost (the amount of effort involved in receiving the service) and psychological cost (to what extent the customer is feeling uncertain about service outcome versus the expected service). The total benefit is what the customer gains of the service compared to another option in terms of the price (the cost saving compared another actor), the functional (how the service is superior to another option) and psychological (how well the different service options satisfies the customer). Furthermore, having a good cost of service also have a positive effect of the perceived quality of the service (Abu-Salim, et al., 2017).
2.3.3 Service Delivery
affected by to what extent the customer can rely on the deliver i.e. that the all of ordered goods are delivered on the desired time (Lai & Yang, 2009). The service level of the delivery is further dependent on the overall logistics service and performance. Low error rates and cargo damage leads to an increased logistics performance as well as an increased service performance (Lan, et al., 2016).
Further, the service level on the delivery is affected by the ordering handling system at the company. The service level is positively affected by a rapid response time on orders which can be enhanced by providing the order service with adequate data and information in combination with a good communication and information exchange system (Lan, et al., 2016; Seuring & Goldbach, 2002).
2.3.4 Flexibility
Flexibility has commonly been defined as a company’s ability to respond to new situations or market changes. In logistics this translates to the company’s ability to quickly respond to the customer needs in inbound or outbound transport, which may include rerouting transports or adapting to new delivery times. When it comes to the flexibility of logistics four areas have been identified that affect the overall performance of the logistics flexibility: flexibility of physical supply, flexibility of purchasing, flexibility of physical distribution and flexibility of demand management (Maldonado-Guzman, et al., 2017; Shah & Sharma, 2014).
Physical flexibility is the company’s ability to quickly and efficiently provide a variety of incoming materials and suppliers. Purchasing flexibility is the company’s ability to buy a variety of materials and make agreements with different suppliers. The physical distribution flexibility is the company’s capability to adjust warehouse levels, packing of products and transports quickly to meet the customer demand. Demand management flexibility is the firm’s ability to respond to the variety of customer service needs and delivery times quickly. Further, a good demand management can help a company to differentiate themselves from a competitor. From a customer’s viewpoint the physical distribution flexibility and demand management flexibility are valued highest since they are most visible and relevant for the customer. However, to successfully achieve that an efficient flexibility of physical supply and purchasing is necessary. (Shah & Sharma, 2014)
2.3.5 Green Supply Chain and customer involvement
Green supply chain management (GSCM) is becoming increasingly important as customers become more aware of the environmental impact of the goods they purchase. “GSCM refers to
the intra- and inter-firm management of the upstream and downstream supply chain aimed at minimizing the overall environmental impact of both the forward and reverse flows” (Chavez,
et al., 2016, p. 207). Shibin et al. (2016) describes key enablers for a GSCM Financial stability,
logistics optimization and strategic outsourcing are some of the important factors to make a
GSCM possible.
Financial stability is important since implementation of green initiatives often require funding
environmental posts, (Shibin, et al., 2016). Further, it is argued that green initiatives will result in financial benefits in the long run (Srivastava, 2007).
Logistics optimization is positively related with environmental performance with regards to
greenhouse gas emissions. Optimizing the load utilization, speed and routing are options to consider when increasing the logistics optimization (Shibin, et al., 2016). Further, to increase the logistics optimization it is important to evaluate the network design for the distribution and all of its ingoing parts (Srivastava, 2007).
Strategic outsourcing can help an organization to save costs on their logistics and will enable
an increased focus on its core competences. The use of 3PL will ultimately increase the flexibility of the distribution and allows for more frequent shipping while reducing the overall fixed costs in the distribution (Shibin, et al., 2016).
Stakeholders in the supply chain, especially the end customers, can affect the adoption towards a more environmental supply chain. This causes firms to increase their environmental quality and performance. This is the case especially in western countries that large customers pressures firms to increase their environmental performance which increases the motivation for a cooperation between the customer and supplier to implement more environmental practices. Understanding the customer needs is an essential part of the GSCM. Customer centric green supply chain management is a strategy that can be used to better understand the customer needs on the environmental quality of the supply chain. Customer centric GSCM involves both the customer and the organization to mutually plan, set targets and management practices for a GSCM. (Chavez, et al., 2016)
Overall, a successful customer centric GSCM will have a positive effect on the operational performance in quality, cost, delivery and flexibility which will increase the overall customer satisfaction (Chavez, et al., 2016).
2.4 Demand Management
Figure 3 - The Demand Management Process (Croxton, et al., 2002) 2.4.1 Collect data & Information
There are several ways to make a forecast and depending on what type of forecast different data will be required, for products with low demand variation and high volume it is suiting to use a forecast based on historical data. For products with high variation in demand the forecast should consist of human input. To collect all the data in question many different parts of the organizations may have to be contacted (Croxton, et al., 2002). Regardless of what type of data that will be required a company to make a forecast, Chopra & Meindl (2016) presents six factors that a company need to consider when making a forecast:
• The past demand • Lead time
• Future planned activities
• Planned discounts and sale events • State of the economy
• Actions that the competitors have taken 2.4.2 Forecasting
products aggregated but disaggregated per region (Croxton, et al., 2002). Generally a forecast decline in accuracy the further up in the supply chain it is and the longer the span it covers (Chopra & Meindl, 2016).
There are several methods for making a demand forecasting, Rushton et al. (2014) presents four different methods that can be used: Judgemental methods, Experimental methods, Causal
methods and Projective methods.
Judgemental methods – Are suitable when there is limited historical data, this type of forecast
should be based on the opinions from the customers as well as senior persons in different departments of the company.
Experimental methods – Methods that should be enhanced when there is no information that
can be used for a forecast, which can be the case for new products. Here companies can try the new product on a small geographic location to test what the demand will look like for a larger population.
Casual Methods – Can be used when the demand is highly dependent on several factors such
as the state of the economy, seasonality, weather, competitors etc. This forecast method finds a correlation between the depending factors and demand and then use future estimates for the depending factors to make a forecast.
Projective methods – also called time series models is a method that uses historical data of the
demand to find trends which can be used to project the demand into the future. This forecast method assumes that the past demand is a good presentation of the future demand, which makes this a good method to use when the demand pattern looks similar from year to year. A common proactive forecast method to use is to plot the past demand in a graph to identify seasonal trends, as well as plotting a trend line over a longer time period to identify if the demand is increasing or decreasing.
2.4.3 Synchronizing
Figure 4 - Synchronizing Process adopted from (Croxton, et al., 2002) 2.4.4 Reduce variability & Increase Flexibility
Demand variability is complicated to plan for, and it is expensive to manage. However, there are two ways of dealing with demand variability. The first is to reduce the variability and the second one is to increase the flexibility. Working with both reducing variability and increasing flexibility are central components of demand management. The first focus should be to reduce the variability since it aids consistent planning and helps in reducing costs. Flexibility is important to be able to meet demand from unpredicted events, but it is usually costly. Demand variability can emerge from several different sources, in the demand management process these sources should be identified as well as ways to reduce them (Croxton, et al., 2002). There are several ways to reduce the variability, a company can adjust the service specification (change lead time, order days), establish standards that limits the offers (Monczka, et al., 2016), set a minimum order quantity that covers all costs or to integrate the demand volatility into the network design (Croxton, et al., 2002).
Further, Chopra and Meindl (2016) suggest that the price can be adjusted depending on the time to steer the demand. For instance, if a company want a more even demand curve, the price can be lowered during low sales periods and increased during peak periods. However, successfully eliminating all demand volatility is seldom the case, so a company need to have flexibility to meet the volatile demand. The level of flexibility needs to be determined based on the financial and supply chain capabilities of the company and the customer needs (Croxton, et al., 2002). To increase the flexibility a firm can consider strategic outsourcing and ensure appropriate capacity levels in the supply chain for volatile demand (Monczka, et al., 2016).
2.4.5 Measure performance
3 Method
This chapter describes the choice of methodology and research design of the study. Further, the quality of research is presented, assessing the validity, reliability, and generalizability of the research. Lastly, the ethical considerations of the research are presented.
3.1 Choice of Methodology
The research methodology of this report has followed the case study approach suggested by Yin (2003). The use of a case study in research is appropriate when trying to understand a contemporary phenomenon over which the investigator has limited or non-existent control (Yin, 2003), which is the reality of this study. Another reason for the use of a case study is that the researcher seeks to answer a “how” question (Yin, 2018). The case study is of exploratory nature since it aims to create an understanding of the underlying causes of the research problem.
3.2 Research Design
This research has followed the process illustrated in Figure 5. The initial stage of a context study provided an overview of the problem area and helped guide the decisions regarding data collection, and literature to review. The context study is described in a separate chapter below. The empirical and theoretical parts of the research process have been conducted in parallel with an early focus on the theoretical aspect to form the theoretical framework. During the work, focus has shifted toward the empirical aspect with the collection and analysis of data, aided by the theoretical framework. As an extension of the data analysis, scenarios of how plausible actions could affect the process were developed. Finally, the results were discussed and summarised as recommendations for possible actions and their effects.
Figure 5 – Research Process 3.2.1 Choice of theoretical framework
From knowledge gained through the context study, research areas and relevant literature fit to serve as a theoretical framework were identified and has included the following:
1. Distribution networks
4. Demand management
In order to find academic literature within the identified areas, Web of science, Google Scholar and KTH primo were used as primary search engines for scientific publications. In the search process, key words were identified and used in the search for literature. The main key words were: FMCG, customer satisfaction, supply chain management, distribution networks, cross-docks, collaborative supply chains, demand management, supply chain collaboration. Further, only publications in English and Swedish were considered. The relevance of the articles was determined based on the number of citations, journal publisher and the year of publication.
3.2.2 Context Study
In qualitative research, the researchers need to consider the context of the real-world problem being studied (Korstjens & Moser, 2017). As part of this study, the researchers attended meetings involving the main stakeholders in the collaborative distribution network with the intention of receiving a contextual overview of the problem area. The observations of the context study had a purpose of contributing to the contextual understanding of the problem. As the understanding of context is of great importance in the process of formulating a research problem, the context study was initialized at an early stage in the research process. Apart from attendance at meetings, the observations have included visits at both the central warehouse of the case company as well as at one of the cross-docks in the distribution network, see Table 2. These observations provided knowledge in how operations are handled in the distribution network. Insights from the context study has also aided the researchers in the choice of relevant concepts and literature to review.
Table 2 - Observations conducted in the context study
Observation type Description Date Duration
Meeting Initial meeting with the case company where the problem area was described
2019-10-14 1h
Meeting Meeting with representatives from Pernod Ricard, Carlsberg, and Altia.
2019-12-05 2h
Warehouse observation
Observation of the central warehouse. Product flow and operations were studied.
2020-01-09 4h
Meeting Meeting with representatives from Pernod Ricard, Carlsberg, and Altia.
2020-01-09 2h
Cross-dock observation
Observation of the operations at a cross-dock.
2020-01-22 2h
Meeting Meeting with representatives from Pernod Ricard, Carlsberg, and Altia.
3.2.3 Data collection
The data collection has consisted of two main parts. One part being the qualitative data from interviews and the other one being the part regarding quantitative data in the form of Excel documents. The quantitative data has mainly been used to answer RQ1 as it included relevant volume data to identify and understand shortcomings in the distribution toward cross-docks. The qualitative part of the data collection has had the purpose of highlighting shortcomings, but also to enlighten the researchers of what possible actions that could be taken from a practical perspective, answering RQ2. Quantitative data from one cross-dock was used for conducting an in-depth analysis to assist answering RQ1 and RQ2. The cross-dock in Trollhättan was chosen for an in-depth analysis. It was argued to be representative for all the cross-docks, with an annual volume close to the average annual volume for cross-docks in medium and minor cities. The distribution of shipments above and below the desired shipment volume was also representative for the average of all the cross-docks.
The collection of qualitative data has been in the form of interviews with, and observations of, relevant individuals objects and connected to the collaborative distribution network. The interviews have followed a semi-structured template with open ended questions. Semi-structured interviews have a purpose of drawing forth the individual view and opinions of the interviewee (Cresswell, 2013). The interviewees and the objects of observations were chosen to include perspectives from the different stakeholders in the distribution network. The selection was further aided by representatives from the case company with insights in who could be appropriate to interview, and what observations that could be useful in the study. Interview respondents included representatives in the areas of logistic operations, supply chain management, customer service, sales, and information management.
Table 3 - List of conducted interviews
The theoretical framework was used as a basis in forming the interview questions and structure. Initial questions were directed to understand the respondent’s role in the individual organization as well as in the collaboration. These initial questions provided information in terms of experience and responsibility of the respondent. Following the initial questions, concepts from the theoretical framework assisted in creating relevant questions for the study. From the concept of supply chain collaboration, questions regarding information sharing and decision synchronisation were derived. To achieve an understanding of the customer needs and the provided service level, the literature regarding customer satisfaction provided guidance. Demand management was used to form questions regarding how demand is analysed and what actions are taken to steer and synchronise the demand. The observations in this study has included both visits at the central warehouse and at a cross-dock, as well as attendance at physical and online meetings with representatives from the actors in the distribution network.
The main source of numerical data has consisted of Excel files including cross-dock delivery data of volumes and dates of delivery. Files were received from an actor in the distribution network, and the data spanned from week four of the year 2019 to week three of 2020. The data contained volumes delivered to the different cross-docks with specification of product and date. In order to achieve valid results from the data it was processed and checked for errors as suggested by Saunders et al. (2016). In this study errors from large deviation in form volume
Respondent Title Company Conducted
R1, R2 Wholesales Manager (R1), Customer Service Specialist (R2)
Pernod Ricard 2020-01-27 Face to face R3, R4 Distribution Manager (R3), Outbound
Transport Coordinator (R4)
Altia 2020-01-29
Face to face
R5 Key Account Manager SCM Carlsberg 2020-02-04
Face to face
R6 Key Account Manager On-trade Carlsberg 2020-03-30
Online
R7 EDI manager Carlsberg 2020-03-31
Online
R8 Head of delivery planning Carlsberg 2020-04-02
Online R9, R5 Key Account Manager Supply Chain
(R9), Key Account Manager Supply Chain (R5)
spikes caused by events were investigated and removed if it was not a reoccurring event. The volumes of different products were accumulated to a daily total volume with the purpose to receive overview of the delivered volumes to each cross-dock. Further, cross-docks that had been closed in 2019 or were planned to be close in 2020 were removed from the spreadsheet.
Further data regarding the order stop time and lead time for the customers, shipment time from the central warehouse and arrival time at the cross-dock were received from a partner in the distribution network. Data from a customer survey conducted in 2017 was received from a partner in the distribution network. The survey consisted of different ranking questions with answers from 230 customers connected to the collaborative ordering system. The questions regarded the importance of certain service elements and was answered on a scale from one to four, with one being not important and four being very important.
3.2.4 Data analysis
The results have been presented according to the different categories of the theoretical framework to provide with structure and clarity. The analysis of numerical data was conducted in Microsoft Excel. When analysing the collected data, an exploratory data analysis approach was used to find different connections and patterns. The approach suggests the usage of graphs and charts to achieve a deeper understanding when analysing the data (Saunders, et al., 2016). After the collected data had been sorted through i.e., removing data causing errors, week numbers and weekdays were added next to the date field in each spreadsheet for each cross-dock. Afterwards, the data was presented in a matrix pivot-table to get a good overview of the delivered volume to each cross dock in terms of both weekly and daily delivered volume, Table 4 displays the matrix outline. Week number is indexed by i and the weekday is indexed by j.
Table 4 – Pivot matrix
Week Mon Tue Wed Thu Fri Weekly total sum
3 Vol3,1 Vol3,2 … … Vol3,5 ∑ Vol3,j
4 … … … ∑
. … … … ∑
. … … … ∑
52 … … … ∑
Daily total sum ∑ Voli,1 ∑ Voli,2 ∑ Voli,3 ∑ Voli,4 ∑ Voli,5 ∑∑Voli,j
requirements were presented in a pie chart to get a good overview of the problem as suggest by Saunders et al. (2016). Further, Trollhättan the cross-dock that represents the problem well, were analysed on a deeper level to provide a comprehensive view of the problem. The pivot table for Trollhättan was used to calculate the total number of shipments, percentage of shipments in different volume spans and the number of deliveries to customer and cross-dock per week. The Excel count.if command was used to calculate the total number of shipments as well as the number of shipments within different volume spans. The number of shipments was then calculated to the percentage of shipments within the different spans by dividing it with the total number of shipments. The percentages within the different categories were then displayed in bar charts.
To analyse the cost and earnings depending on the volume that were shipped, an average margin per litre and average stop cost provided by Pernod Ricard, those number were however adjusted due to confidentiality requirements. The calculation for the cost per delivered litre is presented in equation 1 and the earnings per delivered litre is presented in equation 2, and the total earnings are presented in equation 3. For the conduced scenarios, an average earning per delivery was calculated by using the average volume per shipment and the earnings per delivered litre, see equation 4 and 5.
𝐶𝑜𝑠𝑡 𝑝𝑒𝑟 𝑑𝑒𝑙𝑖𝑣𝑒𝑟𝑒𝑑 𝑙𝑖𝑡𝑟𝑒 = 𝑆𝑡𝑜𝑝 𝑐𝑜𝑠𝑡 𝐷𝑒𝑙𝑖𝑣𝑒𝑟𝑒𝑑𝑣𝑜𝑙𝑢𝑚𝑒 (1) 𝐸𝑎𝑟𝑛𝑖𝑛𝑔𝑠 𝑝𝑒𝑟 𝑑𝑒𝑙𝑖𝑣𝑒𝑟𝑒𝑑 𝑙𝑖𝑡𝑟𝑒 = 𝑀𝑎𝑟𝑔𝑖𝑛 𝑝𝑒𝑟 𝑙𝑖𝑡𝑟𝑒 ×𝐷𝑒𝑙𝑖𝑣𝑒𝑟𝑒𝑑𝑣𝑜𝑙𝑢𝑚𝑒 − 𝐶𝑜𝑠𝑡 𝑝𝑒𝑟 𝑑𝑒𝑙𝑖𝑣𝑒𝑟𝑒𝑑 𝑙𝑖𝑡𝑟𝑒 (2) 𝑇𝑜𝑡𝑎𝑙 𝑒𝑎𝑟𝑛𝑖𝑛𝑔𝑠 =𝐸𝑎𝑟𝑛𝑖𝑛𝑔𝑠 𝑝𝑒𝑟 𝑙𝑖𝑡𝑟𝑒 ×𝐷𝑒𝑙𝑖𝑣𝑒𝑟𝑒𝑑𝑣𝑜𝑙𝑢𝑚𝑒 (3) 𝐴𝑣𝑎𝑟𝑎𝑔𝑒 𝑑𝑒𝑙𝑖𝑣𝑒𝑟𝑒𝑑 𝑣𝑜𝑙𝑢𝑚𝑒 =𝑇𝑜𝑡𝑎𝑙 𝐷𝑒𝑙𝑖𝑣𝑒𝑟𝑒𝑑 𝑉𝑜𝑙𝑢𝑚𝑒 𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑠𝑡𝑜𝑝𝑠 (4) 𝐴𝑣𝑎𝑟𝑎𝑔𝑒 𝑒𝑎𝑟𝑛𝑖𝑛𝑔𝑠 =𝐸𝑎𝑟𝑛𝑖𝑛𝑔𝑠 𝑝𝑒𝑟 𝑙𝑖𝑡𝑟𝑒 × 𝐴𝑣𝑎𝑟𝑎𝑔𝑒 𝐷𝑒𝑙𝑖𝑣𝑒𝑟𝑒𝑑𝑣𝑜𝑙𝑢𝑚𝑒 (5)
The conducted interviews with recorded audio were transcribed. As for the interviews that were not recorded, notes were taken during the interviews and were compiled in a text document after the interview. The analysis of this qualitative data was built upon the concepts described in the theoretical framework. Hence, statements and episodes of the data was categorized based on which of the four concepts they related to. If a statement could be categorised in two or more of the concepts the researchers jointly decided on what the main theme of the statement was and categorized it accordingly. After the categorisation of the data, each category was analysed to identify shortcomings related to the distribution towards cross-docks. These shortcomings were then listed in their respective category and described in the results chapter.
options in the aggregation were considered and the final decision of using the three scenarios was based on the scenarios being plausible in the operational context. The direct consequences of the scenarios are described in the results chapter, whereas the possible internal and external effects and implications are further discussed in the discussion chapter.
3.3 Quality of research
The quality of the research can be evaluated in terms of four different aspects: external validity, internal validity, construct validity and reliability (Gibbert, et al., 2008) where internal validity has been excluded from this evaluation since it is not relevant in exploratory research.
3.3.1 External Validity
External validity refers to how well the findings of the study can be generalised outside of the specific case. Blomkvist & Hallin suggest that there is low generalisability on a case study since it can be too specific (Blomkvist & Hallin, 2015). Therefore, this study should not be considered as generalisable in the whole area of supply chain and distribution. However, it could be used to provide empiric examples of shortcomings in certain parts of cross-dock distribution and how they might be addressed. Such a generalisation could be applied in another context of FMCG where low shipment volumes are experienced, using the same concepts on a different study object.
3.3.2 Construct Validity
The construct validity can be described as to what extent the study investigates what it says it will investigate. From that viewpoint there is reason to argue for this study to have an acceptable level of construct validity. The study has investigated what is stated in the purpose. However, in terms of comparing and verifying qualities of results with similar studies the work has been limited. The reason for this is that the specific methods of quantitative data analysis used by the researchers in this study has not been found in comparable research in a similar context.
3.3.3 Reliability
The reliability of this research refers to the consistency of the results and to which degree the study is replicable. The quantitative part of this study can be considered to have a relatively high reliability. This is motivated by the data set spanning from week three of 2019 to week three of 2020. If this study were to be replicated, with a data set from a different year, the results of the quantitative analysis would most likely be similar. However, the scenarios would be difficult to replicate accurately since they rely on the researchers’ perception and interpretation to some extent. Assessing the consistency in data measurement, there is a need emphasise that data has not been collected by the researchers but received as data files from the case company. Despite this, the researchers have no reason to believe that the data is anything but true.
responses have been a central part of this study in providing insights from individuals involved in the process.
3.4 Ethical considerations
4 Case Background
This chapter gives a background to the case company, their situation, and a description regarding the collaborative ordering system. It also describes the main stakeholders, previous events and the current situation in the collaboration, relevant to understand the context of the research.
Pernod Ricard is one of the world’s leading wine & spirits companies and their goods are retailed worldwide. In Sweden, their business is separated into two segments, on-trade and off-trade. The on-trade customer segment includes restaurants, hotels and bars while the off-trade customer segment includes wholesales and Systembolaget. For on-trade customers, direct orders are made possible by a strategic collaboration with Carlsberg Sverige, were they offer restaurant customers their combined portfolio including wine, spirits, beer and soft drinks. This is beneficial for restaurants since they can order a wide range of products and receive it in one delivery, with one invoice. The collaboration is called Drink Link and was launched in 2004 with the aim to attract more customers and achieve a more efficient distribution. Blinge & Svensson (2006) investigated the collaboration and stated that it would increase the utilization of the trucks used for the distribution.
Up until recently, Carlsberg owned and managed logistic assets in the form of personnel, trucks, and facilities. Logistic operations such as goods distribution and transport were mainly handled by the own organisation. These operations also included the distribution of Pernod Ricard products, on- and off-trade, from the cross-docks to customer. However, a strategic decision made by Carlsberg in 2016 changed this, as the discontinuance of these logistic operations was initialised. The decision meant that operations would successively be shut down and outsourced. The decision was made on the grounds of reducing risk (R6) and this transition was finalised in the spring of 2020. Today, all operations regarding Carlsberg’s distribution of goods are outsourced to the company X-Sam which handles the purchases of distribution transport and services as well as cross-dock operations. The transport of Drink Link shipments from cross-dock to customer is still included in the collaboration agreement as Carlsberg’s responsibility, which is now handled by X-Sam. Table 5 displays an overview of the responsibilities of the main stakeholders in the on-trade distribution network.
Table 5 - Company responsibilities in the distribution network Company Responsibility in the distribution
Carlsberg Sverige Owner of the Drink Link ordering system. Managing a major part of customer interaction in the collaboration.
Altia Sweden Managing warehouse operations of Pernod Ricard. Coordinating the linehaul transport between central warehouse and cross-dock through 3PL actors.
As a part of this transition, Pernod Ricard had to make a strategic decision regarding their off-trade flow. The options were either to choose X-Sam as the distributor of both segment flows from cross-dock to customer, or to find another solution. The choice fell on another solution which had strategic benefits for the off-trade flow. In the current situation, which was realised in the spring of 2019, the Drink Link flow and the off-trade flow of Pernod Ricard products are conducted in two separate cross-dock networks.
Altia is the company responsible for managing the central warehouse for Pernod Ricard. Further, Altia provides Pernod Ricard with the service of coordinating the outsourcing of distribution operations in the off-trade flow as well as the distribution between the central warehouse and the cross-docks in the on-trade flow. The transportation from cross-docks to customers in the on-trade flow is managed by X-Sam. Therefore, the collaborative distribution benefits of the collaboration are only experienced in the last mile distribution, i.e. from the cross-docks to the on-trade customers.
In 2017 Carlsberg conducted a customer survey answered by 230 customers connected to the collaborative ordering system. The survey regarded several service elements related to the distribution. Table 6 shows the results from the survey, displaying the different service elements to the left, and the ranking to the right. The answers were made on a scale from one to four with one being not important and four being very important. The table reveals that the most important service elements is with regards to the quality and delivery of the service.
Table 6 –Summary of results from a customer survey conducted in 2017 (Internal Data) Service item Importance of the service item 1 to 4. 1 2 3 4
Suppliers pick up empties. 2% 1% 6% 91%
That there are no shortages on the ordered goods. 2% 2% 7% 89%
That the supplier takes back empty pallets. 5% 3% 8% 84%
That late shipments are announced. 1% 6% 14% 79%
That shipments are delivered to the desired position inside the restaurant.
4% 5% 14% 77%
That there is an option to receive an extra shipment if needed. 5% 10% 18% 67%
That the customer can choose day of shipment, Mon – Fri. 13% 16% 30% 41%
5 Results & Analysis
In this chapter, the results and analysis are presented. To align the results with the research questions, focus has revolved around shortcomings in the distribution towards cross-docks. The main shortcomings identified were low pallet utilization, high costs related to the linehaul, synchronization of delivery days to customer. This chapter provides an overview of the main shortcomings, assumptions, and facts related to the studied case as well as other factors that might be of influence. The structure of this chapter was adapted from the theoretical framework with an additional part regarding the scenarios.
5.1 Distribution Network
To visualize the current state of the on-trade distribution, Figure 6 was developed based on information received from the context study (R3). It shows the physical flow of products from the central warehouse to the customers i.e., restaurants. The distribution network is configured in a hub-and-spoke manner, with one central warehouse distributing to customers via 17 cross-docks allocated throughout Sweden.
Figure 6 - Distribution network configuration
Central warehouse – At this location, the majority of Pernod Ricard’s product volume is stored.
The central warehouse operations are managed by Altia Sweden which is another actor in the wine and spirits market. At the warehouse, physical operations mainly include handling inbound and outbound truck shipments and order picking (R3, R4). The order picking is done with the help of electric pallet trucks. Scanners and barcodes are used to minimize picking errors. Smaller orders, destined to the same cross dock, can be loaded onto the same pallet.
Linehaul – Altia Sweden also provides a service of administering the transportation of
shipments to the cross-docks. This is done through linehaul transportation where different haulage contractors are considered depending on their usual routes and geographical reach in Sweden. Some cross-dock shipments are loaded onto the same linehaul truck if the truck capacity and loci of destination allows for it. The haulage contractors also consolidate other goods from other industries (R3).
network which is managed by X-Sam. Shipments from Pernod Ricard and Carlsberg arriving at the cross-dock are consolidated and loaded onto pallets depending on the location of end customer. The consolidation of shipments is done in a manual way using printed lists describing product type, quantity, and destination. The cross-docks also consolidates other types of goods from other actors and industries. Figure 7 shows the geographical spread of the cross docks as well as the locaion of the central warehouse.
Figure 7 - Geographical locations of cross-docks and central warehouse
Distribution – From the cross-docks, the consolidated shipments are distributed through milk
rounds where a truck makes stops at several customers in a predetermined route (R5). This distribution is managed by transport carriers operating in the cross-dock’s allocated area.
truck makes, a stop cost is incurred. The stop cost is independent of the weight and volume of the goods delivered. Secondly, the cost of a pallet spot on the truck, which slightly varies depending on the haulage contractor. Of these two costs, the stop cost is the greatest, usually 4 to 8 times the amount of the cost per pallet (R3).
Each cross-dock has a set order cut-off time. This order cut-off time acts as a deadline for placing orders for customers connected to the cross-dock. Orders placed before the cut-off time are consolidated to be delivered within the order lead time. For the majority of the cross-docks, lead time has a maximum limit of 48 hours, with the exception of some cross-docks with larger volumes where the lead time is 24 hours (R3, R5). Depending on the geographical location of the cross-dock and the order lead time, shipments are delivered to the final customer either the same day as the shipment arrives at the cross-dock or the following day. Cross-dock locations, order cut-off times and cross-dock delivery times are presented in Appendix 2.
5.2 Customer Service Level
One of the major issues in the distribution is the high number of shipments made and several of them having a low volume. This is partly because there is a service requirement from some of the customers. Depending on the size and importance of the customer, the customer can choose when and how often they will receive delivery. From the interview with R5 it was found that the current service level provided to customer allows some of them to select when their delivery day should be and, depending on their geographical location, the number of delivery days that should be available. However, the service demand from the customer depends on the geographical location. For instance, restaurants in the major cities (Stockholm, Göteborg and Malmö) require a higher level of service in terms of choosing delivery days and the number of delivery days, while restaurants in the less populated areas might be satisfied if there exist more than one option for delivery days (R5, R9).
Restaurants often have a limited storage space in their facilities highlighting the importance of service elements such as correct order deliveries and the returning of empty pallets (R8). Restaurants also tend to have limited economic funds to place larger orders to keep a certain stock level of beverages and therefore rely on restocking with weekly shipments, making restaurants dependent on that there are no shortages (R5, R6, R8, R9). Table 6 further reveals that with regards to flexibility, it is more important for the customers to receive an extra delivery if needed than selecting their regular days for delivery.
