The basic indicators of performance for the transit corridors are the cost and time. How much and long it is required to transport cargo from a point of origin to its final destination. Additionally to these two earlier mentioned indicators, three other indicators of which the quality of services, the infrastructure and the traffic flow of cargo have been assessed (Arnold et al., 2005) [8] . The authors assert that corridor quality is measured in terms of reliability (that is the diversity of services offered), transit time and cost ensured by shipment. In its researches, Degbe S.A. (2017) [9] asserts that the shipment of goods perspective also relies on cost and time but with a focus on the supply chain management. Its studies further suggested that performance can be enhanced by reducing the time and cost at specific nodes within the corridor. The quality of services is the utmost critical indicators for a transport corridor performance and hence for the effectiveness of the corridor nodes. The principal inputs into such indicators are usually time and cost which helps in recognizing the components of a transport corridor that would offer the greatest savings if they are improved.

With the rapidly emerging transport networks, the choice of a chain, rather than a port, becomes more critical for shipping companies, logistics service providers and shippers (Notteboom T.E., & De Langen P.W., 2016) [10] . Scholars have been advocating investigating port choice in a global setting considering the total transport chain instead of just one of its components. Throughout gathered literature, the apparently two main decision variables influencing shippers in the choice of port to call to, are the costs and the efficiency in transport supply chain servicing. The land-leg is often more costly than ocean transport and becomes a differentiating factor in port selection. Still, Tom V. & Cathy M. (2016) [11] found that cost and land transport systems and service levels appeared to play a dominant role in the port selection process.

The attraction of a port to its hinterland relates to the port capacity and mainly to its hinterland transport cost. As noted by Michele A., et al. (2017) [12] ports competing for throughput and of which hosted countries share common hinterland markets are only as strong as their weakest link in the door-to-door transport chain. Tom V. & Cathy M. (2016) [11] stated that given the significant transport cost of the unimodal transport, shippers can be expected to adapt their logistics operating model and lower their service level requirements employing transport combinations lead by barge or rail. As a cost-friendly, the combination of transport modes is greatly favorable for the medium and long distance. The transport decision concentrates on choosing between logistic pathways using different maritime gateways. Instead of a matter of transport mode choice, the transport decision is characterized as a “routing” decision, as Tom V. & Cathy M. (2016) [11] noted. A survey among Austrian shippers and forwarders, conducted by De Langen P.W. (2007) [13] also found that they were not attached to one particular port or mode of transport and could easily switch their routing behavior, especially as the commitment to use combined transport stems from the hard cost savings (Feo-Valero M., et al., 2011) [14] . Their survey concluded that in the medium and long distance, intermodal transport alternatives can leverage economies of scale and reduce their cost. This allows overruling the classic transport economics on the basis of distance advantage.

The comparative cost analysis presented by Cazzaniga Francesetti D. (2005) [15] of routing containers through the Italian ports instead of the major load centres demonstrated the idea that port competition comes into play through the expensive inland leg. Accordingly, port competition becomes a function of the overall network performance. Whereas ports choice is analyzed following the rationale of supply chain managers searching for faster and more cost-effective ways to transport their goods from door-to-door. The analyses of revealed preferences by Malchow M., & Kanafani A. (2001) [16] , and Malchow M., & Kanafani A. (2004) [17] confirmed that port selection by American shippers was submitted in the overall strength of the total transport chain, as they were rather looking for the most performing solution.

Scholars of the field, like McGinnis M.A. (1979) [18] and Cullinane K., & Toy N. (2000) [19] , have been generally concentrating their study efforts on measuring the sensitivity to the cost, speed, reliability, risk for losses and damages of the transport service. Meanwhile, Tsamboulas D.A., &. Kapros S. (2000) [20] identified three generic buyer groups of these transport services. The first category of buyers was referred to as intensive users of intermodal transportation. This category was flexible with service levels and could decide almost exclusively in function of cost. The second group was more stringent as it combined cost and quality criteria when deciding on intermodal transport options. A smaller third group hardly used intermodal transport given particularities such as refrigerated transport.

The analysis by shippers of the choice between routing alternatives confirms the findings of the many pieces of literature indicating that cost is leading. Tom V. & Cathy M. (2016) [11] proved that even minor transport chain cost differences of less than 1.5%, an equivalent of approximatively 20 euros in their research, prevailed in the transport decision among shippers. Several pieces of research have succeeded in establishing a relationship between reduced logistics costs and trade growth. Few authors to mention, Hummels D., et al. (1999) [21] estimated that exporters in East Asia who were able to achieve a 1% reduction in their shipping costs could realize a 5% to 8% increase in their market shares. Almost identically, a later study of operational efficiency in the South Asia Subregional Economic Cooperation (SASEC) corridor used a gravity model based on tariff reduction to estimate that a 1% reduction in transport costs would produce a 5% expansion in trade according to (Padeco Consultants, 2004) [22] .

Several contributions can be found on freight routing and costs minimization, many of which advocate transport intermodal as means of land-leg transport efficiency in terms of costs minimization. The minimization of land-leg transport costs is the main goal of the models presented in the literature (Rizk N., et al., 2006) [23] ; (Selim H., et al., 2008) [24] ; (Lee Y.H. & Kim S.H., 2000) [25] ; (Chern C.C. & Hsieh J.S., 2007) [26] ; (Jang Y.J., et al., 2002) [27] . The post haulage in the intermodal transport chain is performed by road and the cost curve of intermodal freight transport thus runs parallel to unimodal road transport. Once the total intermodal cost is calculated, it is possible to make comparisons with unimodal transport. This opens the way to a series of possible scenarios that can be assessed using an appropriate set of tools according to (Cathy M., et al., 2011) [28] .

Barnhart C., & Ratliff H.D. (1993) [29] considered the problem of determining the minimum cost routing for each shipment with the combination of truck and rail. The costs included transportation and inventory holding costs. Likewise, Boardman B.S., et al. (1997) [30] considered the problem of determining the transportation route and the combination of transportation modes while minimizing cost and time.

The data used in this study, are obtained from various West African countries Chamber of Commerce and the PMAWCA (2016) [39] . Table 2 shows the cargo demand in million tons from the 3 LLCs. The data information on the corridors distances (see Table 4) and the average transport cost (see Table 5) are respectively obtained from Google map, and the Borderless. The research will refer to the basic indicators for evaluating the corridor performance of five countries,

Source: Burkinabe, Mali and Niger’s chamber of commerce and PMAWCA (2016) [39] .

from their respective port (seen as the point of origin) to the LLCs commercial cities they serve.

The research forecast the cargo volume flow for the next 10 years (from 2017 to 2026). To attain this forecast, the researcher proceeds by simply calculating the rate of evolution by variation period of n and then the overall average evolution rate which is both mathematically presented as followed:

Calculation of Rate Evolution by variation of year,

t G = V j − V i V i × 100 (1)

t G : Rate of evolution by variation period of (n).

V i : Initial volume of year (n).

V j : Final volume of year (n + 1).

Calculation of Overall Average Evolution Rate,

T m = t G 1 + t G 2 + ⋯ + t G n n (2)

T m : Overall Average Evolution Rate.

t G 1 : Rate of evolution by variation period from (n) to (n + 1).

t G n : Rate of evolution by variation period over (n) period.

n: number of variation period.

From the precedents mathematical calculations, we obtain the forecasted results of the cargo demand in million tons from the 3 LLCs, presented in Table 3 which starts from the year 2017. Therefore this study is using the year 2023 as a benchmark.

Consideration is given to the transport system between an origin and destination (OD) pair in the model. The study assumes that the traffic demand is generated from major commercial cities (capital cities) of each of the three LLCs. This, due to the fact of unpublished traffic volume from other potential important commercial city centres of the studied LLCs. The 15 alternatives available corridor routes, involve three LLCs, Burkina Faso, Mali, and Niger, and five major ports, including the ports of Abidjan, Cotonou, Lagos, Lomé, and Tema (Figure 1).

Source: Processed by the author.

Instead of transport mode comparison, we are studying the selection of alternative corridors to route the containers by using different ports. A route is considered as from a point origin to a point of destination (OD) irrespective of the transport mode.

Table 4 summarizes the distance of the corridors between the five respective ports (Origin) and the major commercial cities of the LLCs (Destination).

A comparison overview of the distance from O to D is given in Figure 2. The longest corridors to and from Mali, are respectively the Cotonou-Bamako, Lagos-Bamako, Lomé-Bamako, and the Tema Bamako. Abidjan is at the shortest distance away from Bamako destination with about 1382 km. Regarding Niger, Lagos-Niamey appears to be the shortest path, closely followed by the Cotonou-Niamey route, then the Lomé-Niamey corridor route. The Lomé-Ouagadougou corridor, is the shortest path among all other four corridors routes to and from Burkina Faso. The corridor route of Tema-Ouagadougou is the second, and followed in third position by the Lagos-Ouagadougou corridor route. Abidjan appears to be the most distant to Ouagadougou with 1232 km. Analyzing port choices in the United States, Blonigen B., & Wilson W. (2006) [41] concluded that distance and transport prices and individual port efficiency play a significant role in determining a port’s market share. Figure 2 illustrates the length of the corridors from origin to destination.

The average transport costs in USD per ton per kilometer are represented in Table 5. This average transport price mechanism is calculated based on the fuel cost, road conditions, as well as any indirect and direct costs associated with inland transportation. Among the cheapest road transport rate, is Ghana’s with 0.14 USD/ton/km followed by Togo, 0.16 USD/ton/km. The highest transport rate is noted with Nigeria, 0.2 USD/ton/km. The objective is the identification of the minimal transport cost route. Hence in the following course of the corridors

Source: Google map [40] .

Source: Borderless (2016) [42] .

transport cost performance analysis, where direct rail transport is available and operational, the model will make calculation utilizing the average rail transport cost instead of the average road transport cost. In the case of Abidjan to Ouagadougou, different costs are allocated to different transport modes. The transport modes considered are two namely; rail and truck.

A Questionnaire was systematically administered to a chosen target population at the Port of Lome. In all, 100 questionnaires were dispatched for the survey of which 70 copies were retrieved; and only 60 copies were considered valid for the survey analysis. Information was gathered from several interviews and that data gathered was used to derive current conditions from the general perception in

the transport supply chain.

This illustrates the approach used by the researcher in selecting the respondents for the case study. Random sampling was used to choose a select group of the population. This sampling method is meant to be an unbiased representation where every member stood an equal chance of being selected for sampling. It has been noted by Descombe (1998) [43] that when selecting a sample size, there are two major factors that can be explained about the population and sample size in qualitative research. Firstly, they are unlikely to be known with precision or certainty at the beginning of a research paper. Secondly, the sample size will generally be relatively small to the entire population. Hence, an adequate and comprehensive data must be obtained in order to provide a solution the outlined problems.

The questionnaire therefore mainly focused on the first stakeholders along the Lomé-Cinkassé transit corridor and extending to Ouagadougou, Niamey and Bamako. The target population subject to the survey was the freight forwarders, the truck operators, and the transit importers and exporters. The questionnaire was accompanied with an introductive letter of the researcher and its research purpose in Appendix 1. To this utilized method was added the ID Number attributed to each respondent, to actually ensure not only the anonymity of the respondent’s information, but also to avoid duplicate respondents’ survey. This survey technique also assures a higher response rate from surveyed participants and encourage their openness. A sample of the questionnaire can be found in Appendix 2. The results and analysis of the survey are presented in the following sections.

The total transportation calculation cost can simply be written mathematically as follow:

T T C = ∑ n = 1 k d n D n F n (3)

where,

TTC = Total Transportation Cost.

d n = distance between two locations.

D n = quantity in millions of tons.

F n = average transportation cost in USD ton per kilometer.

The unimodal transport cost for freight transiting from each of the five major ports to the destination to each of the three LLCs commercial centres (Bamako, Niamey, and Ouagadougou) is presented in the Table 6. This freight rate comparison only considers the amount paid to the railways or the truck and does not include terminal handling costs nor loading or unloading cargo at final delivery.

By the year 2023, the volume of cargo imports and exports to and from Bamako (Mali) would increase by 128.9 % from the year 2015, a yearly growth rate of about 10.9%. The best option route in the case where coastal countries by 2023 continue to still use a unimodal mode of transport for shipping cargo, would be first the Abidjan with USD 3.472 Billion. This is followed by Tema

Source: Processed by the author.

USD 4.07 Billion and in the third position the Lomé corridor route USD 4.666 Billion (see Figure 3).

In regards to the cargo to and from Niamey (Niger) to the coastal countries, the first option route shippers might consider would likely be Cotonou USD 2.302 Billion, followed by Lomé USD 2.337 Billion and then the Lagos route USD 2.457 Billion in terms of unimodal transport costs. Abidjan and Tema are the most costly transport corridor routes among others (see Figure 4).

Though the distance between Lomé-Ouagadougou is listed as the shortest compared to other routes, Burkinabe shippers would likely in the forecasting unimodal transport cost prefer the Tema route costing USD 2.144 Billion. The Lomé-Ouagadougou route, USD 2.152 Billion would come as the second best

corridor route option, followed by the Abidjan-Ouagadougou route USD 2.499 Billion. This is interesting given that cargo routed through the Abidjan-Ouagadougou corridor goes by train (see Figure 5).

In regard to the introduction of an intermodal transport system along the Lomé corridor route, 55% of the respondents opposed the idea with the majority 53% of the truckers, who argued that such transport modes would increase the rate of unemployment in the transport industry (see Table 7). On the contrary, 42% of respondents welcomed the idea of an intermodal transport system; of that 22% of the truckers and 20% of freight forwarders who stated that this would lead to transit time saving, security improvements, avoidance of the Alédjo-Mountains, transport cost reduction, and the reduction of road accidents.

Considering the choice of the intermodal terminal along the Togo corridor route, the majority of the respondents’ for the adoption of an intermodal transport system, here about 36% voted in favor of the city of Mango as the best host city for an intermodal terminal (see Table 8).

Source: Processed by the Author.

Source: Processed by the Author.

In the previous section, the survey-based terminal localization gives the city of Mango, at 550 km north to the capital city of Lomé, as to be the best intermodal terminal location. This research, therefore, uses the survey results for calculating the intermodal transport cost of cargo flow through the Togo corridor. Calculations are thus taking into account this new parameter. The transport cost is a function of several factors such as the volume of cargo, the distance from the origin to final destination, the average transport cost by road and by rail, the demand from each LLCs, and the available mode of transport.

The choice of the best route in terms of cost minimization will depend on the total transport cost whether in unimodal or intermodal case. The total unimodal versus intermodal transport cost calculation is presented in Table 9. The section deals with analyzing transport costs by routing. This implies the comparison of the different cargo traffic flow from ports and routed to each LLCs market. The upper limit on the freight rates will be fixed by the road freight rates; i.e. the railway cannot set the freight rates higher than the road freight rates unless the customer is willing to pay more than the road freight rates. As for Rajiv D., et al. (2011) [44] , they suggest that the freight rate should be set in sort to determine the optimum pricing policy and terminal location in order to guarantee a certain level of profit for the railway from the intermodal service. This means that pricing should be set in such a way that the railway can achieve profits without losing traffic to road transport.

Moreover, for the case of Togo, it is given based on our survey research that customers are not willing to pay more than the road freight rate, and therefore the truck shipping cost using the intermodal service should be less than the cost of shipping by road. Consequently, the research applies the freight rate of 0.14 USD/ton/km as same as the one from Ivory Coast, added to the fact that these two countries share numerous similarities.

Regarding the forecasted demand, the Ivory Coast corridor offers the least inland transport cost for cargo traffic in direction and from Bamako in Mali, with a total cost of USD 3.472 Billion. This followed by the Tema-Bamako route, with about USD 4.070 Billion then the Lomé-Bamako route with USD 4.503 Billion. The Benin and Nigeria corridors offer the highest transport costs USD 5.179 Billion and USD 5.740 Billion, respectively (see Figure 6).

Source: Processed by the Author.

The Togo corridor route is considered to offer the least transport cost with USD 2.205 Billion for cargo in direction to and from Niamey in Niger. The Benin and Nigeria corridors come as second and third with USD 2.301 Billion and USD 2.457 Billion respectively. Abidjan-Niamey route exhibits the highest transport cost, therefore, cannot be considered unfortunately as profitable for Niger shippers (see Figure 7).

In regard to the transit cargo to and from Ouagadougou in Burkina Faso, Togo corridor route offers the best transport cost in terms of cost minimization with USD 1.992 Billion, directly followed by Ghana corridor exhibiting a transport cost of USD 2.144 Billion and the Ivory Coast with USD 2.499 Billion. The Lagos-Ouagadougou routing holds the highest transport cost among all the routes, with the Cotonou-Ouagadougou routing closely behind (see Figure 8).

The city of Mango intermodal terminal and the application of the intermodal transport for transit cargo to and from the 3 LLCs through the Togo corridor, would enable substantial transport cost savings for shippers. This will, therefore, reflect on the final product price in the retail stores.

A shift from unimodal transport to the intermodal transport would by 2023 lead to a transport cost saving USD 162.58 Million on the Lomé-Bamako routing. The Lomé-Niamey routing transport cost saving would account for USD 131.47 Million while costing savings the Lomé-Ouagadougou routing USD 159.4 Million. An overall transport costs savings of USD 453.45 Million for the 3 LLCs understudy in the account of the year 2023 (see Figure 9).

As a result, the costs advantage to serve medium and long distance with rail is significant and is directly reflected in the pricing of intermodal transport. The

rail-truck intermodal service can be a key strategy adopted to shift the high traffic density from the road to the railway. It will include the advantages of both the truck and the rail service. Pricing and willingness to pay are fundamental in the strategic planning for the development of the new infrastructure for an intermodal transport system.

The present research assumes that road distances will not alter within the time period forecasted. An average transport cost per ton-kilometer remains the same. It also assumes that transport costs are the main factors intervening in the transport decision, as we can realistically suppose that operators try to optimize the transport mechanisms. Other factors which have not been taken into account in this model, although they should not be neglected in the final decision, include speed, security of transport, deterioration of the goods and queuing time. The study found that the function of transport costs is most important in the transport intermodal problem, and determine the choice of cargo routing by shippers.