Keywords

1 Introduction

Agri-food supply chains (AF-SCs) consists of organizations that are responsible for the production, distribution and sale of vegetable or animal-based products. In general, two main types of AF-SCs exist (van der Vorst 2000):

  1. 1.

    SCs for fresh agricultural products. Normally, these SCs are formed by growers, auctions, wholesalers, importers and exporters, retailers and speciality shops. The most important processes that these actors perform in the SC are the planting, harvesting, handling, storing, cleaning, sorting, packing, transportation, and especially trading of these goods. Examples of these SCs may be fresh vegetables, fruit or flowers.

  2. 2.

    SCs for processed food products. These SCs may comprise the same actors than the previous case, but with a special importance for the processors. They usually take the agricultural products as raw materials for producing products with higher added value. Some examples of these SCs may be desserts, snacks or canned food products.

Modern AF-SC involve millions of actors with different objectives and interests, organized in multiple stages, and continuously taking a great variety of decisions requiring complex, multilateral, and multilevel governance. For instance, there are several million farmers, several hundred thousand different food processors and retailers, and 500 million consumers in the EU. The figures are much larger if the total number of global agents who are involved in the EU AF-SC are taken into account (Bachev 2013).

This sector, complex in itself, is further complicated by the existence of high levels of uncertainty. According to van der Vorst (2000), SC uncertainty refers to “decision making situations in the SC in which the decision-maker lacks effective control actions or is unable to accurately predict the impact of possible control actions on system behavior”. It may occur because of a lack of information or understanding of the environment or also, because of a lack of a consistent model of the SC that include the relationships between SC variables and performance indicators.

van der Vorst (2000) affirms that one of the most important aspects in SCs is the uncertainty reduction in decision making. This uncertainty complicates or even may prevents that the logistical objectives of the SC can be reached, or that the customer orders can be achieved, for example, due to unexpected changes in the weather conditions that could have a huge impact on the quantity produce by the SC.

The management of high levels of uncertainty remains a major concern because of its high impact on the AF-SC decisions and its processes, despite the fact that, according to van der Vorst (2000), by breaking down the walls that are present between successive SC stages, decision making uncertainties may decline, since more information and control possibilities will become available to the decision-makers in each stage. In this line, Mor et al. (2015) conclude, from the comprehensive literature assessment, that the safety, quality and associated economic benefits in sustainable AF-SCs can be achieved through elimination of uncertainties, in addition to innovation, supply chain collaboration and introducing global supply chain practices into lean and green initiatives.

In this context, as a tool for the uncertainty analysis in the AF-SC, a framework of analysis is proposed to evaluate the impact of the sources of uncertainty on three key aspects (quantity, quality and time) for each of the stages of AF-SC and at each decision level (strategic, tactical and operational).

The remainder of this paper is organized as follows; in Sect. 2 the agri-food Supply Chain characteristics is described. Next, Sect. 3 addresses the decisions in the agri-food Supply Chain, establishing three decisional levels for each AF-SC stage. In Sect. 4 are listed the uncertainty sources and a possible classification of them. Section 5 contains a framework to quantify uncertainty sources impact in the agri-food Supply Chain and some quantification impact examples. The paper concludes by summarizing the main results.

2 The Agri-Food Supply Chain Characteristics

It is a widely accepted fact that the agri-food sector has a number of special features that make it especially complex. van der Vorst (2000) summarizes a list of specific process and product characteristics of AF-SCs from other papers and categorizes it by stage in the SC. Thus, in the Growers and Producers, it should be noted long production throughput times and seasonality in production, whereas in the Auctions/Wholesalers/Retailers we can find variability of quality and quantity of supply of farm-based inputs, seasonal supply of products requires global (year-round) sourcing, and requirements for transportation and storage means.

In general, according this author, the Food industry can be characterized by: (a) uncertainty about the quality and quantity of products supplied from farms, (b) production systems oriented to the volume instead the product variety, (c) systems focused on capacity utilization because of highly sophisticated capital-intensive machinery used, (d) high variability in the process yield, regarding quantity and quality obtained, due to biological variations, (e) high seasonality and existence of random factors as the pests, weather and other biological hazards that can act simultaneously, (f) alternative installations, cleaning and processing times highly product-dependent, (g) importance of the quality in all parts for that the agricultural inputs can be complementary, (h) importance of lot traceability of the product to control the work in process and to prevent quality problems, (i) necessity to storage the material, intermediates or finished products in containers or special tanks with special conditions, so can appear problems with the buffer capacity or bottlenecks, (j) importance of the recycling of materials and (k) decay for raw materials, intermediates and finished products that decrease the quality products along the process.

From a point of view of process, Verdouw et al. (2010) identify the basic trans-formations (or stages) in a generic fruit supply chain:

  • Growing and harvesting: is the first stage and consist of the production of ripened fruit from inputs as seeds, trees, soil, water, fertilizers and pesticides. The basic activities are planting, pruning, thinning, fertilization, irrigation, crop protection and crop maintenance, and picking fruits from trees.

  • Washing, sorting and grading: after harvesting, the fruits are usually washed in order to clean the earth and insects, and then sorted and graded according to its size, color, weight, ripeness or shape.

  • Processing: consist in the transformation of fresh fruits into food products such as juices, jams, muesli or ice creams.

  • Packaging and labeling: to facilitate its manipulation, storage and transportation, the fruit is packaged. It is very usual to add it also a label with product-related information that may be needed for further processing or for the customer, such as description, variety name, origin, harvesting date, product code, ingredients, etc.

  • Storage and distribution: according to demand and taking into account the decay the fruit is receipted, warehoused, dispatched and transported to the wholesalers or to the retailers.

  • Retailing: in the last stage, the fruit is delivered to the final consumers in specialized fruit and vegetables shops, supermarkets, markets in the street, restaurants, etc.

3 Decisions in the Agri-Food Supply Chain

Decision making in organizations usually follows a hierarchical approach. This mean that the decisions made at one certain level are constrained by those previously taken at the higher level, and besides, constraining those must be taken at the next lower level. Therefore, to understand and manage the organizational decision making processes, it will be very important understanding the interactions among managerial or hierarchical levels (Silver 1991). On the other hand, to take the best decisions, it will be necessary to consider several aspects such as profile of the decision maker, planning horizon, level of detail of the input information and uncertainty associated, and frequency of decision making. In accordance with Ganeshan et al. (1999) three levels of logistical management can be distinguished:

  • Strategic management: At this level the organizational goals and the strategies for attaining these goals are defined. These decisions are mainly concerned with the establishment of the SC configuration.

  • Tactical planning: At the second level the organizational goals and market performance demands are translated into logistical objectives. Tactical planning reflects decisions for the coming weeks or months. These decisions include the choices on and implementation of information systems and organization structures to be discussed in the next sections.

  • Operational control: This managerial level is concerned with the daily operation of a facility to ensure that the most profitable way to fulfill actual order requirements is considered and executed. It contains all operational decisions, which directly influence the flow of materials or information. Typically, operational decisions reflect day-to-day operations up to two weeks in advance.

Taking this into account, it will be possible to structure the decisions in each AF-SC basic stage, discussed in previous section, according these three decision levels. In this way, we’ll have a strategic level, a tactical level and an operational level for the growing stage, and the same for each stage in the AF-SC. In Table 1, it is shown the resulting matrix when crossing decision levels with AF-SC stages. Each quadrant has been denominated with a code formed by the first letter of the corresponding decision level (S-Strategic, T-Tactical, and O-Operational) and a correlative number for the SC stages (1–7).

Table 1 Decisions levels for each AF-SC stage
Full size table

In this way, some important decisions in each quadrant could be:

S1: Crops location, crops quantity.

T1: Month/week to plant seeds (the same decision for each sub-stage according previous section: growing includes activities such as planting, pruning, thinning, fertilization, irrigation, crop protection and crop maintenance).

O1: Week/Day to plant seeds (the same decision for each sub-stage according previous section: growing includes activities such as planting, pruning, thinning, fertilization, irrigation, crop protection and crop maintenance).

S2: Season in which you want to harvest.

T2: Month/week to harvest, quantity to harvest per week en each crop.

O2: Week/Day to harvest, quantity to harvest per day in each crop.

4 Uncertainty Sources and Classification

According to van der Vorst (2000), “sources of SC uncertainty are inherent characteristics of the SC and characteristics of the managed system, managing system, information system and/or organization structure that are present at a certain point in time and that generate SC uncertainty”.

van der Vorst et al. (1998) identify four main uncertainty sources (US): (a) the first and main cluster of sources of uncertainty is the total order forecast horizon, (b) the second cluster of sources of uncertainty is related to input data available for a decision, (c) the third cluster refers to administrative and decision processes, and (d) the last cluster of sources of uncertainty in (especially food) supply chains is inherent uncertainty in demand, process and supply.

Other US’s are: weather, plagues, natural disasters (fire, flood,…), diseases (plants and livestock), seeds quality, effects of fertilizers, ripening pace, agricultural machinery breakdowns, waste rate, product quality decay influences shelf life, product loss at transport/distribution stage because of accidents, delivery time overdue because of traffic congestion or vehicle breakdown, machine breakdowns, scrap rate, strikes, illness of staff, changes in laws and regulations, changes in trade agreements between countries or regions, new descriptions or technological/biological advances, customer demand and competitors influence.

On the other hand, van Donselaar (1989) identifies four types of uncertainty in the SC:

  1. 1.

    Demand uncertainty, associated to the customer’s orders.

  2. 2.

    Supply uncertainty, taking into account the delivery of raw materials in time, according to the customer specifications and in the right amount.

  3. 3.

    Process uncertainty, inherent to the production system and related to the machinery breakdowns or the manufacturing and setup times.

  4. 4.

    Planning and control uncertainty, related, for example, to the difficulties to know accurately inventory levels, with to make a right forecasting, with the fulfilling delivery times or with the correct communication about the customer requirements.

These types of uncertainty sources allow us reordering the above uncertainty sources in the following way (Tables 2 and 3).

Table 2 Uncertainty sources classified by type
Full size table
Table 3 Connection between Decisions levels for each AF-SC stage table (Table 1) and uncertainty sources classified by type table (Table 2)
Full size table

This US classification can be easily connected with the above table (Table 1) about decision levels and AF-SC stages due to Supply, Process and Demand are conceptually like successive stages in a SC. In this way, it can be drawn the following relations.

5 Quantifying Uncertainty Sources Impact in the Agri-Food Supply Chain

Regardless of whether an US may be more or less important at any given time, or at any stage, there are three aspects in the SC that can be affected: quantity, quality or time aspects (van der Vorst 2000). Hence, each US could be analyzed according the amount of impact (high ↑, low ↓ or none “ ”) it exerts on each aspect (quantity = QN, quality = QL and time) of each AF-SC stage and in each decisional level, as shown in the table, for the fruit producer stage at operational level:

 

Fruit producer

 

Growing

Harvesting

Distribution

Strategic

S1

S2

S3

Tactical

T1

T2

T3

Operational

O1

O2

O3

 

Supply

Weather

QN

QL

TIME

QN

QL

TIME

QN

QL

TIME

↑

↑

↑

↑

↑

↑

  

↑

Plagues

QN

QL

TIME

QN

QL

TIME

QN

QL

TIME

↑

↑

 

↑

↑

    

Natural disasters

QN

QL

TIME

QN

QL

TIME

QN

QL

TIME

↑

↓

 

↑

↓

   

↑

Diseases

QN

QL

TIME

QN

QL

TIME

QN

QL

TIME

↑

↑

↓

↑

↑

↓

  

↓

Seeds quality

QN

QL

TIME

QN

QL

TIME

QN

QL

TIME

↓

↑

↓

↓

↑

↓

  

↓

Effects of fertilizers

QN

QL

TIME

QN

QL

TIME

QN

QL

TIME

↑

↑

↑

↑

↑

↑

  

↓

Ripening pace

QN

QL

TIME

QN

QL

TIME

QN

QL

TIME

  

↑

 

↓

↑

  

↓

Agric.machin. breakdown

QN

QL

TIME

QN

QL

TIME

QN

QL

TIME

     

↑

  

↓

Inherent uncertainty

QN

QL

TIME

QN

QL

TIME

QN

QL

TIME

↑

↑

↑

↑

↑

↑

  

↓

So, we could say, for example, that the seeds quality can have a low impact in the amount of product (quantity) that will grow and will be harvested, a high impact in the quality of product that will grow and will be harvested, and a low impact in the time of grow and will be harvested. On the other hand, the seeds quality won’t have any impact in the quantity or quality of distribution, while it could have a low impact in the time the distribution takes place.

6 Conclusions

In this paper, we have enumerated several US’s in the AF-SC and we have presented a framework to evaluate the impact of the US’s on three key aspects (quantity, quality and time) for each of the stages of AF-SC and at each decision level (strategic, tactical and operational). The impact evaluation of the US allows to better know the importance of each US and, hence, to define specifics actions trying to reduce this uncertainty.