Keywords

1 Introduction

Logistics allows coordination between procurement, production and distribution [1]. Companies with world-class logistics capabilities achieve a competitive advantage by providing a superior service to their customers [2]. Competition requires the delivery of the appropriate product timely, in the right conditions and at the lowest cost, for which logistics capabilities are necessary [3]. The correct management of logistics capabilities generates operational efficiency, productivity and an increased value for the customer [4, 5]. [6] Identify three policies for planning logistics capabilities: An overcapacity to cover unforeseen demands; a capacity used in its totality even if demand is not met and, Ensuring that capacity and demand coincide. These logistical capabilities can be internal (organizational) or external (in SC) [4]. In the two large logistical processes, storage and transport, there is qualitative and quantitative logistical capacity. The quantitative capacity of storage or transport refers to the maximum number of cargo units that can be stored or transported [7]. Capacity planning and management minimizes the gap between the necessary capacity and the current one [8] in order to mitigate risks within the supply chain (SC). Companies have the choice of expanding logistics capacity through their own infrastructure or through the contracting of 3PL [9]. The interest in logistics outsourcing is evidenced by the increase of companies that use logistic operators (3PL) to manage their logistics operations [10, 11] as it allows a reduction of costs, greater responsiveness, better customer service and decreased assets [12]. The importance of 3PL derives from global markets, produced goods are transported over large distances, stored for varying periods of time at different locations along the supply chain (SC) [13].

In perishable food supply chains (PFSCs), logistics management requires specific elements. The capacity must allow food preservation and speed in logistics processes in order to reduce losses. The organization must respond to unexpected situations related to logistical requirements, while still preserving quality [14]. In the PFSCs there are high requirements for qualitative logistics capabilities, given the use of vehicles and warehouses equipped with conservation technologies [15]. While in developed countries post-harvest food losses are low, in developing countries they are high due to a lack of adequate transportation, storage facilities, packaging, and handling techniques [16, 17]. The cold chain (CC) is a factor to consider in the management of qualitative logistic capacities in PFSCs, the shelf life of the food can be affected by temperature conditions [18]. A controlled temperature maintains the quality of the food [19]. Cold storage and transport is essential, it preserves quality and reduces the growth rate of microorganisms that cause damage [20, 21]. This method of conservation in PFSCs, for example for fruits or vegetables, allows the preservation of the organoleptic characteristics and keeps the products in a good state [22, 23].

In this article, we propose a model for the management of the logistics capacities for PFSCs in a dynamic environment, focused on storage and transport and the evaluation of the expansion of own capacity in comparison to contracting a 3PL, in two scenarios: with and without a cold chain. The model that evaluates the dynamics of the mango SC in Cundinamarca-Bogotá contemplates the discrepancy between supply and demand. In the model, the qualitative capacity is developed by means of a scenario where the warehousing and the vehicles, count or not with system of cold; On the other hand the quantitative capacity is posed as the available capacity expressed in square meters and vehicles by each one of the actors. The article is organized in two parts: first the model for capacity management of the PFSC that includes the dynamic problem and the Forrester model, and in a second part, the application of the model on the SC of mango presenting the chain and the results. In the end, there is a conclusion and a presentation of possible future works.

2 Model for the Management of Logistics Capabilities of the PFSC

Below, the dynamics of the chain and the model designed for the management of capacities of the PFSC is presented.

2.1 Dynamic Problem Description

The dynamics of the PFSC are represented by causal diagrams, which can be reinforcing (positive) or balancing (negative) [24]. The problem studied led to the proposition of the following hypothesis: “The management of storage and transport capacities allows a greater approximation to the actual conditions of the PFSC by not considering capacity as an infinite variable, while the use of a cold chain system supports a decrease in levels of losses as well as the quantity missing and compliance in deliveries.” In Fig. 1, the causal diagram is presented. It is made up by the feedback cycles, logistics capacity, responsiveness, costs and quality.

Fig. 1.
figure 1

Causal diagram of the logistic capacities of mango

Full size image

The quality cycle conformed by R1 and B2, is about control measured in terms of losses. This is in turn linked to consumer satisfaction due to a higher quality in fresh or processed food. It is also affected positively when the cold system is implemented as it helps to preserve quality in transportation and storage. The logistical capacity cycle (R2) takes into account the required capacity and measures it against the available capacity in order to establish the capacity missing in terms of inventory and transport. In the case of the cold chain the level of inventory is expanded when the capacity. The cost cycle (B3): the higher the uses of cold system the higher are the capacity costs, lowering the budget. A higher budget allows for an expansion of capacity, which results in more cold system. The cycle of responsiveness (B4): based on a higher demand, more perishable and processed food is generated, which leads to a greater requirement of warehouses and transportation, reducing the available capacities, influencing the capacity of response and this in turn on the shortcomings that have a negative impact on demand.

2.2 Forrester Model for Supply Chain

A model was developed in System Dynamics using iThink 9.1.3 software. The PFSC consists of farmers, agroindustry, distributors, shopkeeper, hypermarkets and consumers [1]. Figure 2 shows the structure of the chain of food flows (blue arrows) and information flows (red arrows). This model is based on the models made by [1, 18, 25]. This model includes new agents in form of hypermarket, shopkeeper and a logistic operator that supplies transport capacity and warehousing, Fig. 2. In addition, several other modifications are included as presented below.

Fig. 2.
figure 2

Forrester diagram for the agro-industrial supply chain (Color figure online)

Full size image

Variable discrepancy: In Eq. (1), when the supply of food is greater than the demand, the variable takes a positive value, in the opposite case it is negative, if they are equal, it will equal zero. This will affect the inventory level in warehousing.

$$ Discrepancy \, = Food\_available\_of\_the\_agent{ - }Demand $$
(1)

Percentage of food for direct transport (% DT): Based on the food inventory level and the variable discrepancy, either the quantity to be stored or the quantity to be sent for direct transport in order to satisfy the demand is obtained (Eq. 2)

$$ \% DT = IF\left( {Discrepancia < = 0} \right)\;THEN\left( 1 \right),ELSE\left( {Demand/Quantity\_net\_available} \right) $$
(2)

When deciding what amount of food will be stored and what quantity will be transported directly, the model determines the amount of food to be stored in the own warehouse and the amount to be stored by the logistic operator.

Something similar occurs with transport when agents recur to 3PL vehicles after having exhausted its own capacity (Fig. 3). Figure 3 shows the farmer’s use of own and subcontracted storage capacity. The formula (3) is used to determine the quantity of fruit that will be available in the own warehouse. The quantity that is stored in the own warehouse is restricted to the available capacity in square meters. The same procedure is carried out in the case of transport capacity. When there is no more own capacity the operator is used.

$$ \begin{array}{*{20}l} {Food\;in\;inventory\;in\;the\;farmer's\;own\;warehouse = MIN} \hfill \\ {((Capacity\_of\_the\_warehouse\_of\_the\_farmer} \hfill \\ {*\;Converter\_in\_square\_meter),\;food\_in\_inventory\_of\_the\_farmer)} \hfill \\ \end{array} $$
(3)
Fig. 3.
figure 3

Storage with own capacity

Full size image

Discrepancy between food in transit and the demand (DTD) (4): the food leaves the warehouses because of perishability and to satisfy the demand, the latter is given by this variable. When the discrepancy is positive it gives the amount of food that will leave the stores to meet the demand.

$$ DTD = Demand{-}rate\_of\_food\_in\_transit $$
(4)

The model has the power to decide whether to expand the actor’s own capacity or to continue subcontracting, for which the model follows a decision algorithm, Fig. 4. The algorithm is used for capacity expansion of warehouse or transport. In warehouse, extra square meters are acquired and in transport extra vehicles. The model will expand capacity according to the agent’s available budget.

Fig. 4.
figure 4

Decision to expand capacity algorithm.

Full size image

Scenarios: The model allows a simulation of the PFSC with and without the use of cold storage and cold transport system. The scenarios generate a different percentage of storage and transport losses in each of the PFSC agents, including 3PL. PFSC with CC will have half the losses compared to the other scenario, according to [26] the cold doubles the shelf life of perishable foods. In Fig. 5, Forrester is presented in order to calculate the losses of food in storage (LS), the percentage of losses by decay (% LD) and management (% LM), which will depend on the food.

Fig. 5.
figure 5

Losses storage

Full size image

In addition, losses in transport (LT) are calculated. For the change of scenario, the variable “type of system” was included, Fig. 6. From the % LD and % LM the loss rate is calculated with Eq. (5), in which type of system 1 is the scenario not using cold and the scenario 2 is with the cold system. For this reason, when the system type is 2 the rate of losses is multiplied by 0.5. For transport, % LD corresponds to travel time.

$$ \begin{array}{*{20}l} {Rate\;of\;losses\;in\;the\;agent's\;warehouse \, } \hfill \\ { = \, Warehouse\_of\_the\_agent*\left( {\% LD\_agent + \% LM\_agent} \right)} \hfill \\ {*\;\left( {IF\,\left( {type\_of\_system = 1} \right) \, THEN\left( 1 \right) \, ELSE\left( {0.5} \right)} \right)} \hfill \\ \end{array} $$
(5)
Fig. 6.
figure 6

Losses transport

Full size image

3 Application of the Model on the Mango Supply Chain

The experimental design of this research follows the steps of the experimentation of [26], which raises 5 simulations for each proposed scenario, using a different seed value in each simulation. The used values are 1, 2, 3, 4 and 5, a DT = 0.5 was used. The simulation time was 20 years (7300 days) in order to see the long-term effect on logistics capabilities. For the input information of the model, surveys were used with the different actors of the Mango SC (SCM) during the years 2015–2016, using non-probabilistic snowball sampling. 30 surveys were carried out with producers, 10 with wholesalers, 2 with agroindustry, 10 with transporters, 2 with hypermarkets, retailers: 54 with shopkeepers and 40 with market place traders. The primary information was supplemented with reports from AGRONET, FAOSTAT and DANE as well as information from specific research of mango from fieldwork in the producing municipalities of Cundinamarca ASOHOFRUCOL and logistic work on the SC of fruit [27]. This information of the SCM in Cundinamarca-Bogotá allowed for the definition of the production area (ha) and yield (t/ha), seasonality of harvests (months), transit times between links (d), loading and unloading times, consuming population and consumption per capita for fresh fruit and pulp (g-inhabitant/d).

3.1 Fruit Chain: Mango

The mango is a fruit from the inter-tropical zone of juicy and sweet pulp. It is characterized by its good taste and its nutritional content [28]. The use of cold has a positive effect on climacteric fruits like mango, as it decreases the respiratory rate, the speed of biochemical reactions and enzymatic reactions. The CC preserves its quality [29]. With a temperature of 7–12 °C and relative humidity of 90%, a shelf life of 3 to 6 weeks in storage is achieved. The production of mango in Cundinamarca represents 34% of the total national production. It is one of the fruits with greatest demand [27]. In stock, the % LD of the mango will depend on its respiration rate, which varies according to the number of days that has passed. The number of storage days were modelled with a random variable, the LMs are a constant value for each agent, for farmers it is 2,9%, for wholesalers it is 5.6%, for hypermarket 5.76%, for shopkeeper 5.76% and for market places 5.6%, agroindustry 3,5%. For transport the % LD is the one that corresponds to one day as the duration of the trips is very short. The respiration rate is 0,41% in the case of not using a cold system and with a cold system it is 0.2%.

3.2 Results in the Supply Chain of Mango

The quality in the model is measured as a function of the amount of losses. The results are presented in Table 1. The reduction in total losses is 2.9% in the scenario with CC in comparison to without the cold chain. The main losses occur in the farmer’s warehouse and in the 3PL. This is due to the excess supply in the harvesting season, generating an imbalance between supply and demand, increasing for the farmer with cold storage as storage times increase. Although the fruit lost in transport is a small share of the total, when the cold system is implemented, the loss of transport is reduced by 34.03%, as an effect of the all CC on the mango.

Table 1. Losses of transport and storage
Full size table

The response capacity is measured in the model as the percentage of compliance and the percentage of non-compliance. Fruit compliance was 51% without cold system and 73%, with cold system, Fig. 7. Regarding non-compliance, both in fruit and in processed products, the indicator was better with a cold system, improving by 27.64% and 30.61% respectively.

Fig. 7.
figure 7

Demand average compliance percentage of supply chain mango.

Full size image

Sensitivity analyses 1: It was done on losses, increasing the percentage demand and supply. For the case the demand, the losses in transport were increased due to a greater flows food, while in storage a decrease was observed given that there is a greater rotation in the warehouses. On the other hand, by increasing the supply, the losses increased in both transport and storage, since large quantities of fruit did not flow quickly, Tables 2 and 3.

Table 2. Losses of transport and storage by increase percentage demand
Full size table
Table 3. Losses of transport and storage by increase percentage supply
Full size table

Sensitivity analyses 2: It was made for the farmer regarding his utility, for being the actor with less budget has for the expansion of his capacity, although the one that requires it the most. For the without cold scenario, there is an increase in the increased transport and storage capacity, while in the cold scenario, although transport capacity is increased, the storage capacity was not modified due to the high costs, Table 4.

Table 4. Expanded transportations and warehousing capacity of farmer
Full size table

The costs of increasing capacity per agent were measured for the two scenarios, agroindustry had greater investment in both scenarios, it was the only agent that decided to expand its storage capacity, it used more own capacity than subcontracted capacity. The farmer and the wholesaler were the other actors who incurred in the cost of expansion, but in transport. As well as in the agro-industry, the costs of the cold system are greater. The costs of expanding own capacity are lower than those of subcontracting, as there were actors who did not expand their capacities in any of the scenarios as a result of lower subcontracting costs, Table 5.

Table 5. Cost of capacity expansion (millions of pesos).
Full size table

For all the actors, more is always stored with the CC. In transport the opposite occurs, more vehicles are used in the scenario without cold, but only slightly, the difference being between 10 and 15%. When cold is used, the losses are reduced. This results in fewer trips to comply with orders. The farmer and agroindustry are the only actors that use their own transportation capacity. All other actors use the 3PL operator; the wholesaler does not transport fruit to its customers. The farmer and agroindustry expanded more with the CC scenario, by 30.4% and 63.6% respectively. Both are making more use of their own transport capacity than that of the operator. Table 6 shows the behaviour of capacity expansion in quantities, some actors did not expand their own capacity (farmer in storage, wholesalers in transport) because they were less expensive to subcontract to 3PL or perform very low storage, such as cross-docking. The fact that they do not increase their own capacity involved an expense in subcontracting.

Table 6. Expanding their capacity or subcontracting 3PL.
Full size table

4 Conclusions

A management model for logistics capabilities for the perishable food supply chain has been developed, which evaluates the expansion of own capacity in comparison to contracting a 3PL in two scenarios: with and without cold chain. It includes the seasonality of supply and its discrepancy with demand, allows the study of logistical capacity, quality, costs and the responsiveness. Losses in the PFSC can be reduced by the correct utilization of logistics capabilities and the implementation of a cold chain. The farmer is the actor that suffers the greatest amount of losses in transport and storage. Most of this effect happens due to oversupply caused by the harvest seasons, which causes farmers to store large quantities of fruit. The farmer and the agroindustry are the ones that expand their transport capacity the most. Agroindustry is the only actor that expands its storage capacity, gives priority to use its own capacity instead of using the capacity of the operator because it has the budget to do so. The farmer has no budget because his/her profit margin is low.

For future work the efficiency of the 3PL could be evaluated with regards to the agents of the SC comparing own logistic capacity against subcontracting with 3PL.