Juan et al. (2014) add to this definition that the cooperating companies achieve common objectives. Usually, this objective is related to minimizing overall distribution costs (Juan, Faulin, Pérez-Bernabeu, & Jozefowiez, 2014).
Collaboration can be implemented into companies at different kinds of levels. It can be strategic: sharing key infrastructure and/or company sensitive information. An example of infrastructure collaboration or sharing is terminals, pipelines or warehouses. Another meaning of strategic collaboration could be a long-term business contract and the sharing of demand capacity information (Audy et al., 2012). Companies need to determine an accurate value of the collaboration to obtain their sharing strategy. At a strategic level, companies are likely to share all information on demand to get this value as accurate as possible (Frisk, Göthe-Lundgren, Jörnsten, & Rönnqvist, 2010). If companies do not share the complete demand model, companies cannot reach optimal results (Simatupang & Sridharan, 2002).
Bahinipati et al. (2009) present a framework to establish a successful relationship between supply chain partners. Their paper defines the strategic set of decisions as answers to the following questions: who to cooperate with? What is the optimum number of partners? (Bahinipati, Kanda, & Deshmukh, 2009). By defining three similarity parameters these questions can be evaluated: competence similarity, which describes the degree in overlap between cooperating firms; competence similarity, which describes the degree in overlap in geographical coverage between cooperating firms; and cultural similarity, which describes the overlap in culture and management style (Amer & Eltawil, n.d.).
When applying collaborations to an operational level; logistics collaboration, the decisions presented can be supplemented by the following. It is important to decide who will be responsible for what as well as who will own leadership, which information is needed and how benefits are shared. Within a collaboration, it is possible that responsibilities shift from one company to the other to improve or maximize the results of the collaboration (Audy et al., 2012).
Sanchez et al. (2015a) present a framework for horizontal logistics collaboration (HLC). The model identifies four elements that need be considered before embarking on a HLC project: outset considerations, ideal required synergies, assisting enables and output metrics. The outset considerations can be seen as boundary conditions and need to be appraised before starting the project. The ideal required synergies are desirable conditions for the model to work effectively. The assisting enables can improve performance of the model, which can be measured by the output metrics (Sanchez Rodrigues, Harris, & Mason, 2015). The parameters linked to these elements are indicated per element:
• “Outset Consideration Factors: legislation (external factor), trust among partners, LSP support, common supplier and delivery base and competition issues, level of supply chain control of the companies intending to join a partnership and several factors related to the internal processes within the lead company (access to information and data security, commercial approach to HC and level of volume scale).
• Ideal Required Synergies: internal and external trust among the parties involved, common supplier and delivery bases, directional imbalances and the development of consolidation and deconsolidation centres.
• Actioning Enablers: horizontal partnership, effective commercial model, 4PL orchestrator, capable 3PL and physical and ICT network infrastructure.
• Output Metrics: lead time, shipment frequency, inventory cost and levels and CO2.” (Sanchez Rodrigues et al., 2015)
Figure 3 presents the framework on implementing logistics collaborations after the company decided the first two steps: who to cooperate with? What is the optimum number of partners?
FIGURE 3 COLLABORATION FRAMEWORK (SANCHEZ RODRIGUES ET AL., 2015)
HOW IS BACKHAULING INTEGRATED IN SUPPLY CHAIN NETWORKS?
A generalized logistics network operated by a third party logistics (3PL) company is considered. This 3PL is integrated into the supply chain by providing specialized integrated operations, warehousing and transportation services. Since backhauling is part of the vehicle routing problem which is part of the planning of the 3PL, backhauling is integrated into the supply chain network via the 3PL.
Most vehicle routing problems with backhaul considerations assume there is perfect information available regarding demand (Demirel et al., 2010). This is almost never the case. The uncertainty in demand causes for the backhauling problem, namely, the matching of the carrier to cargo or container takes a while. Ongtang & Sirivunnabood (2014) present a generalized logistics network witch backhaul matching opportunity operated by a 3PL company. The network consists of linehaul source nodes, N. Linehaul products are shipped from N to a linehaul destintation for each route on day, D. When the product is delivered, the truck will travel empty to pick up a backhaul product at the backhaul node close to the origin node. Depending on the 3PL company’s logistics planning policy, the backhaul shipment must be collected on or withing day D+T, where T is the dependent parameter. Then the truck returns, loaded with the backhaul products, to the backhaul node near the trucks origin node (Ongtang & Sirivunnabood, 2014).
From this brief framework, it can be concluded that many routes and combinations are possible. Many papers on the topic of VRP and backhauling aim to minimize the number of empty trucks, thus maximize the number of matchings. To be able to do so four main factors should be kept in mind. Location or, the distance between the linehaul node and the backhaul node (“deadhead”) which should not exceed a prefixed value. The time schedule or, the time it takes to find and load cargo for and onto the truck. Contamination between the products in the linehaul and backhaul shipments. And lastly, the vehicle compatibiltiy or, the truck types matched on the linehaul and backhaul should be the same (Ongtang & Sirivunnabood, 2014), (Mckinnon & Ge, 2009).
A study on forest logistics in Scandinavia highlighted that a proper information network is key to achieve higher efficiency (Carlsson & Rö, n.d.). Demirel et al. conclude that imperfect information is almost never considered in these matching models. However, imperfect information has a big impact on the results of the model, because imperfect information implies search and searching needs time and thus waiting costs (Demirel et al., 2010). The increase in waiting costs might cause for the 3PL’s company to decide to return empty.
WHAT PROBLEMS OCCUR WHEN IMPLEMENTING BACKHAULING INTO A SUPPLY CHAIN NETWORK?
It has been acknowledged that the lack of practical relevance of the vehicle routing models hinders the implementation of efficient backhauling. However, several studies highlight barriers or impediments obstructing effective implementation of backhauling. McKinnon & Ge impute the lack effective implementation of backhauling to six elements: priority is given to outbound delivery services, unreliability of collection and delivery operations, inadequate knowledge of available loads, lack of co-ordination between purchasing and logistics departments, incompatibility of vehicles and products and resource cons
traints (Mckinnon & Ge, 2009). Cruijssen et al. (2007) and Krajewska et al. (2008) agree on much of this list; elaborating on this by identifying partner identification, the uneven negotiating positions of partners and the difference of implementation of information communication technology as key barriers. (Krajewska & Kopfer, n.d.)
FIGURE 4 FACTORS CONSTRAINING THE BACKLOADING OF TRUCKS (MCKINNON & GE, 2009)
UNRELIABILITY OF COLLECTION AND DELIVERY OPERATIONS
Due to the waiting time, as discussed earlier, the chance of backhauling operations being delayed can be high. Depending on the nature of the operation, the level of risk increases or decreases. The chance of delay increases when backhaul networks are more complex, i.e. multiple journey legs and multiple load and off-loading at different nodes. McKinnon and Ge (2004) surveyed the frequency of delays in a UK grocery supply chain. They estimated that about 29% of the journey legs was delayed by 45 minutes in 2002. Most of these delays happened at the reception bays of factories, distribution centres and shops (McKinnon & Ge, 2004). From an operational point of view, backhauling is possible and feasible if enough time is incorporated in the schedule. However, slack in schedules is being tightened and failing the arrival in the narrow time windows awaits penalties. Telematics might slightly improve this situation by enabling real-time visibility of the transport operation and improving communication between driver and origin destination.
INADEQUATE KNOWLEDGE OF AVAILABLE LOADS
Within a network of multiple nodes, many backhaul cargo matches are possible. However many of these are missed because of inadequate transparency between companies. 3PL companies and vehicle operators do not get information on available loads for backhauling. In addition, the same study by McKinnon and Ge (2004) showed that 61 % of the backhaul loads acquired came from “word of mouth”.
LACK OF CO-ORDINATION BETWEEN PURCHASING AND LOGISTICS DEPARTMENTS
If responsibility for transport was to be transferred to the purchasing company, coordination of inbound and outbound deliveries is likely to improve. Which would cause for more backloading opportunities because the physical movement of products can be discussed as part of the trade negotiations between companies.
INCOMPATIBILITY OF VEHICLES AND PRODUCTS
Specialised products and vehicles with particular handling instructions can set narrow constraints onto backhauling opportunities. When a vehicle is available to take a backload, the vehicle might be unsuited for the kind of cargo and thus unable to backload.
IS IT POSSIBLE TO INTRODUCE HORIZONTAL COLLABORATION TO IMPROVE BACKHAULING?
The previous chapters highlighted how information sharing systems enable companies to cooperate, work jointly to plan and execute supply chain activities together to pursue goals they would not be able to reach when acting alone. ICT boosts core supply chain processes such as forecasting, production and product development to be more visible, which leads to possibilities to collaborate between supply chain partners.
However, the previous answers to the research questions also showed that information sharing systems lack implementation in the backhauling problem. Can the backhauling problem be improved through collaboration and the enabling factor ICT? This chapter elaborates on three case studies showing how the concept of collaboration could improve backhauling. Respectively, a case study on wood supply in Sweden, factory gate pricing in the UK and pallet networks in the UK will be discussed.
WOOD SUPPLY COLLABORATION IN SWEDEN
Eight forest companies in Sweden were considering coordinated transportation planning and wanted to know the potential of collaborating. In this case, the companies regarded their supply and demand as common. The model used in the study artificially integrated the companies into one. The savings associated with this mode was 14.2%. However, the companies would not agree on equally sharing the savings and decided to base the savings on the share of overall volume. The relative savings ranged from 0.2% to 20% which was a too big of a difference. Several sharing principles were tested in a case study. As a result of this case study three companies started a collaboration. The collaboration in which the companies shared a coordinated planning system. Before the start of each month, the companies shared information on supply and demand with the 3PLs. From there, an integrated plan was set up and returned to the forest companies. This sharing principle was tested for four months and potential savings were 5% to 15% each month.