Analysis of Crowdsourcing Logistics in B2C e-commerce: Costs and Environmental perspective

POLITECNICO DI MILANO School of Industrial and Information Engineering Master of Science Programme in Management Engineering ANALYSIS OF CROWDSOURCING LOGISTICS IN B2C E-COMMERCE: COSTS AND ENVIRONMENTAL PERSPECTIVE Supervisor: Prof. Riccardo Mangiaracina Co-Supervisor: Prof. Angela Tumino Author: David Bardasco San José ID number: 10552166 Academic year: 2015-2016 2 David Bardasco San José Crowdsourcing Logistics in B2C e-Commerce 3 ABSTRACT Currently, the logistics in e-commerce is one of the big problems, especially in the deliveries between the company and the consumer, as result it generates additional costs that make the online service is not all efficient that it could be. Therefore, the present work focuses on the study of a new practice in the B2C e-commerce logistics, Logistic Crowdsourcing, as well as the creation of a model of costs and emissions about this innovative practice. To carry out the study about this practice has been made a literature review and the study of some projects of companies that nowadays are offering this service. Later, also it is created a model to calculate the costs and the emissions of the Crowdsourcing Logistics. Regarding the results, in the analysed articles there are nothing about Crowdsourcing Logistics, that is, there is a gap, even if there are real projects around this practice currently being carried out. This way, it has been possible to define this concept, how this practice is made and the most important players. About the creation of the model and the related simulations, it can say that the results have allow to extract some interesting conclusions, which allow to understand still better this innovative practice. So, this paper aims to offer an alternative to current problems in the B2C ecommerce logistics with the Crowdsourcing Logistics and study the costs and the emissions of some hypothetic scenarios with the model created for this practice. And this way understand better the concept and really if it is a good and optimal solution, alternative or not. Key Words: Crowdsourcing Logistics B2C Crowdsourcing Logistics, B2C Logistics, Crowdsourcing Last Mile, B2C Crowdsourcing. 4 David Bardasco San José CONTENTS ABSTRACT _______________________________________________________ 3 CONTENTS _______________________________________________________ 4 LIST OF FIGURES __________________________________________________ 6 LIST OF TABLES __________________________________________________ 7 INTRODUCTION ___________________________________________________ 8 JUSTIFICATION ...................................................................................................... 9 OBJECTIVE AND SCOPE..................................................................................... 10 1. LITERATURE REVIEW ___________________________________________ 12 1.1. INTRODUCTION AND SCOPE OF THE STUDY ........................................ 12 1.2. METHODOLOGY ......................................................................................... 14 1.3. SUMMARY OF REVIEW AND DISCUSSION ............................................. 16 1.3.1. Main characteristics of the papers examined ........................................... 17 1.3.2. Research methods used .......................................................................... 20 1.3.3. Themes arising from the review ............................................................... 20 1.4. CONCLUSIONS ........................................................................................... 45 2. PROJECTS ___________________________________________________ 47 2.1. IDENTIFY OF THE MAIN PROJECTS ........................................................ 47 2.2. CLASSIFITACION AND SUMMARY OF THE PROJECTS ......................... 48 2.2.1. Basic ........................................................................................................ 49 2.2.2. Crowd ....................................................................................................... 52 2.2.3. Deliveries.................................................................................................. 57 2.2.4. Other Aspects .......................................................................................... 60 2.3. CONCLUSIONS ........................................................................................... 61 3. MODEL ______________________________________________________ 63 3.1. Objective ....................................................................................................... 63 3.2. Context ......................................................................................................... 64 3.3. Scope ........................................................................................................... 66 3.4. Methodology ................................................................................................. 67 3.5. Cost Model ................................................................................................... 68 3.6. Environmental Model .................................................................................... 73 4. ANALYSIS OF THE RESULTS ____________________________________ 77 Crowdsourcing Logistics in B2C e-Commerce 5 4.1. Procedure ..................................................................................................... 77 4.2. Costs: Summary and conclusions ................................................................ 83 4.3. Emissions: Summary and conclusions ......................................................... 86 5. CONCLUSIONS _______________________________________________ 90 6. BIBLIOGRAPHY _______________________________________________ 92 6.1. REFERENCES ............................................................................................. 92 6.2. WEBS ........................................................................................................... 97 ANNEXES ______________________________________________________ 101 1. Results of the simulation .................................................................................. 101 6 David Bardasco San José LIST OF FIGURES Page Figure 1: Phases of the Methodology……………………………………………...15 Figure 2: Diagram of Alternatives………………………………………..………..36 Figure 3: How Crowdsourcing Logistics works…………………………………...62 Figure 4: Main window of the created Model (Excel) …………………….……...78 Figure 5: Datum considered………………………………………………………..79 Figure 6: Basic Scenario………………………….………………………………...81 Figure 7: Cases……………………………………………………………………...82 Figure 8: Costs Case 1……………………………………………………………...85 Figure 9: Costs Case 25………………………………………………….………...85 Figure 10: Emissions Case 1…………………………………….………………...88 Figure 11: Emissions Case 25…………………………………………...………...88 Crowdsourcing Logistics in B2C e-Commerce 7 LIST OF TABLES Page Table 1: Basic Classification………………………………….……………………18 Table 2: Area Classification………………………,,……………………………….21 Table 3: Advantages and Disadvantages of Traditional Shopping Model……..23 Table 4: Advantages and Disadvantages of Online Shopping Model………….27 Table 5: Area Classification (II) ……………………………………………………31 Table 6: Advantages and Disadvantages of Alternatives………………………..44 Table 5: Projects…………………………………………………………………….48 Table 8: Basic Classification of the Projects………………………………………51 Table 9: Crowd Classification of the Projects……………………………………..54 Table 10: Deliveries Classification of the Projects………………………………..58 8 David Bardasco San José INTRODUCTION Currently logistics in e-commerce, especially in B2C (Business to Consumer) is one of the main problems, if not the main, because this involved to the companies high costs and in many cases consumers’ dissatisfaction. It is mainly for this reason that nowadays it is trying to find, and implement new logistics practices in order to find the right solution that achieves the satisfaction by such retailers as part of end customers. In this project has chosen to focus and study one of these new logistic practices that are testing and implementing nowadays, specifically Crowdsourcing Logistics. In such a way, to carry out this work and taking into account the scope of this Master in Management Engineering, it has decided to make a review literature in order to see if there is literature about this innovative logistic practice. And then, to analyse if currently there are real projects that are using this practice, it means companies that provide this logistic service. It’s a given that this practice could be the future in the logistics of e-commerce. So, it has decided to make a cost and environmental model with the objective to study better this new practice. And obviously, afterwards makes simulations with this and extracts conclusions. In conclusion, it can say that the main purpose of this project is to study the practice of Crowdsourcing Logistics through the analysis of literature and the study of real cases, that is companies that currently are operating this logistic service. On the other hand, make a costs and environmental model in order to analyse these two aspects about the practice. Following to analyse the model created, have been studied different scenarios with the finality to extract conclusions about this innovative service. And then of this, it has seen and decided if the model is coherent or not. Crowdsourcing Logistics in B2C e-Commerce 9 JUSTIFICATION This project has been made in order to carry out the Master’s Final Project and to obtain the Master in Management Engineering. The reason has taken to make the project about this topic, mainly arises from four reasons, which are:  First, my affinity around the logistics and different aspects related with this. Also and in relation with this that in the future I would like to work in a position related with this area.  The second is the idea of the project offered for the professors, which I consider that is very interesting and clearly could be the future in the B2C e-commerce logistics. So, this has been a great motivation to carry out this project.  The third reason is because I am a regular user of e-commerce. It means that the problems that currently there are in this business model, I have been able to live and feel these in first person, as well as the consequences of these.  Finally, I believe that the sector of e-commerce is a very powerful and has a great present and future. So, to find a solution for one of its main problems could be vital for the development and future as much the B2C e-commerce as the society. Moreover, the cited practice hasn’t still been made known in lots of countries and neither has taken advantage of this. So, it means that these facts give more possibilities of to increase the use of this practice. 16 David Bardasco San José · Delimiting the field: The papers and the information are limited to the related relation with the stipulated field that is especially with the Crowdsourcing Logisitics practise. And the relation with the literature can be with the logistic or with some of the key words that they have been considered. - PHASE 2: Analysis of the selected literature Following in line with Riccardo Mangiaracina et al. (2015), and keeping the objective of this literature review in mind, the select papers classify it on: 1) Their main characteristics: This is a basic classification (title of the journal where the paper has been published, author, year, first author’s country) and the identification of the research method adopted. 2) The main area about which is about the paper. 3) Different criterions about which the papers talk in relation with the main topic. - PHASE 3: Research gaps and potential areas Once time that have been done the previous phases, it could already identify gaps where literature doesn’t talk and could be future areas of research, study. Therefore , in this sectio will determine those areas where it would be feasible or would be more likely to carry out an investigation in line with the focused objectives and clearly around the determined field. 1.3. SUMMARY OF REVIEW AND DISCUSSION In line with Natarajarathinam et al. (2009), it has done the Table I with the objective of summarize the basic information of every paper and the research method adopted. And aligned with Perego et al. (2011), the papers are chronological ordered with the finality to show the evolution of last mile logistic issues related to B2C e-commerce over time. Crowdsourcing Logistics in B2C e-Commerce 17 1.3.1. Main characteristics of the papers examined Next it can observe the main characteristics of the papers examined: 18 David Bardasco San José Table 6: Basic Classification Crowdsourcing Logistics in B2C e-Commerce 19 The total number of analysed papers has been of 15, which were published in 11 different international scientific journals, with a mean value of 1,4 contributions per journal. The papers were published in different types of journals. Specifically, the 73% in logistics and transportation journals and the other in information and communication technologies journals. About the regions addressed, the number of contributions in which the first author is from Finland is 3 (20%), the same that from United Kingdom that is 3 too. Then, United States of America follows them with 2 publications (13%). The rest of the papers (47%) were written by researchers from other countries, which are: Hong Kong, The Netherlands, Jordan, Belgium, Pakistan, Austria and Germany (all these with only 1 contribution by author of every country). So, talking about the first author’s country it can say that Europe is the continent was written more publications (10 papers), followed by Asia (3 papers) and America (2 papers). Also, it has been observed about the authors that Punakivi, M. is the more present authors in the publications, who is present in 3 of the analysed papers and tried to offer a new alternative to Last Mile Logistic. An in regards to the year of publication, it can be identified two clear periods. From 2001 until 2007, weren’t published lots of contributions about the main topic, specifically a 40%. The increase of publications occurred from 2008 until now (2016), where there are the rest of papers (60%). It should be noted that in the year 2014 were where more publications did, followed by the year 2013, 2008 and 2002 with 2 papers every year. The rest of years only carried out one contribute per year. Too it is important to say that the time required conducting a research study, and for a paper to be written, reviewed, and accepted generates that the year of publication of the papers doesn’t match exactly with the situation of the moment. But it can observe that in the last years there is an increase of the publications, as a result of the important growth of the e-commerce. So, the increase is ascending, because every time this type of business is more used by the society and obviously it causes news problems and solutions, alternatives that are reflected in all this papers. 20 David Bardasco San José 1.3.2. Research methods used The publications are classified and evaluated based on their research methodology. And the main categories used are based on a study by Meixell and Norbis (2008), who identified seven research methods, which are: survey, simulation, interviews, mathematical models, case studies, conceptual models and others. The 40% (6 papers) of the papers reviewed present math models; follow by surveys (20%), simulations (13%) and conceptual models (13%) and the rest, interviews and others, with one paper everyone. 1.3.3. Themes arising from the review Next it can observe the classification about the main areas of the papers examined: Crowdsourcing Logistics in B2C e-Commerce 21 Table 7: Area Classification 22 David Bardasco San José 1.3.3.1. Traditional Shopping Model VS Online Shopping Model There are some analysed papers (3 papers) that talk about the difference between the Traditional Shopping Model and the Online Shopping Model. Specifically, in relation with the main topic of this project which is the logistics. Such as say Brown and Guiffrida (2014), the importance of delivery and the supporting logistical process is well recognised in the operations and supply chain literature. Nowadays every one of these models has a different growth and costs and environment impact in the society. It’s obvious that every of this type of model has his advantages and disadvantages too. Following it will analyse what says the literature about these models. · Traditional Shopping Model: The papers that identify and talk about the Traditional Shopping Model are three. And it should be noted that two of these are connected with the environment topic (n.9 and n.13). The Traditional Shopping Model is an option of distributing the goods and it is the traditional system with supermarkets and retail shops (Aized and Srai, 2015). When the customers buy with this model, they themselves pick up the purchased item from the retailer and self-delivers the item to the home using their own vehicle (Brown and Guiffrida, 2014). So, as tell Edwards et al. (2010), the customer does most of the labour intensive work, because she/he has to order picking and transport the goods at home. And it is obvious that in this case, the purchase implies to go to the physical location. About the traditional supply chain, the goods are delivered to store and has been previously cited, the customer picks the items before taking them at home (Edwards et al., 2010). Crowdsourcing Logistics in B2C e-Commerce 23 Therefore, then of see some characteristics of the Traditional Shop Model, it could say that the main advantages and disadvantages of this are: TRADITIONAL SHOPPING MODEL ADVANTAGES DISADVANTAGES · The customers can see what they are buying and if is necessary they can prove it. · Is better for the company, because this form, it hasn’t delivery costs. · The customers can receive help from the shop assistant. · Is uncomfortable for the customers, because they need to go to the physical shop. · Go to the shop has a cost (time and transport). · Is bad for the environment, because every customer goes to the shop and this generates lots of emissions. Table 8: Advantages and Disadvantages of Traditional Shopping Model · Online Shopping Model: All the analysed papers talk about the Online Shopping Model and every of these contribute different interesting information for the topic of this project. The Online Shopping Model is another option of distributing the goods that is a system with direct to consumer deliveries (Aized and Srai, 2015). Specifically, as say Brown and Guiffrida (2014), in e-commerce, the item is delivered to the customer by the retail seller or by an agent contracted by the seller to provide a home delivery service. So, the customers can buy without visiting the physical location. Or in line with Edwards et al. (2010), the vast majority of online purchases result in the physical movement of a small package (or single item) to an individual address (typically a consumer’s home) by parcel carrier. In general, these deliveries are distributed from local parcel carrier depots and consist of mixed loads in the back of vans. So, in this model the retailers must deliver personalised orders to highly dispersed locations within relatively narrow time window. 24 David Bardasco San José In relation with the environment, the Online Shopping Model has yields environmental benefits, because basically it reduces personal travel demand (Edwards et al., 2010). Xu et al. (2008) say that the ability to fulfil and deliver orders on time could determine an e-tailer’s success. Specifically, one of the notable benefits of online shopping is the convenience and time saving when compared to traditional shopping. It includes other factors such as of ease online payment, home delivery and return procedures that all combine to make Internet shopping more convenient than traditional. The value of the home delivery market has increased rapidly during the last few years. And the most common home delivery model is when the customer can select a time window when the items are delivered (Kämäräinen and Punakivi, 2004). · Supply chain: In line with Kull et al. (2007), the current growth and popularity of e-commerce has affected many consumers’ everyday lives by providing a wide range of choices, more available information and ease of purchasing. And it should be noted that as companies extend supply chains via direct delivery to consumers, supply chain efficiency depends upon the usability of the online ordering system. About the supply chain in the Online Shopping Model, Vanelslander et al. (2013) identifies which types of supply chains are the most commonly be found in the online retailing of grocery items in West Europe. And based on this information, there are three types of commonly employed supply chains that are identified, which are: - Pure player: The companies, which use this model, only dedicate to sell their products by e-commerce. It means they don’t have any physical shop where the customers can go to buy their products, because all the actions are by Internet. Inside this model, the main delivery models are: Crowdsourcing Logistics in B2C e-Commerce 25 o Van delivery: This type of supply chain is used by pure player retailers that carry out the last mile delivery with a dedicated fleet of vehicles. The picking operations are performed in one or more dedicated distribution centres, as well as in providing the dedicated vehicles often places a financial burden on the company employing this type of supply chain setup. And in this model, the cost of last mile delivery is generally very high, because of not having sufficient customer density, as well as having to deal with time-slotted deliveries that are very common with this type of supply chain to provide a high customer service. o Parcel delivery: This model employs a parcel carrier’s (e.g. UPS, DHL, etc.) from existing distribution network to deliver the goods to the shopper. This enables the retailer to take advantage of a distribution network that converse a large area and already handles large volumes, thus leading to lower last mile delivery costs. Because this type of delivery to the shopper is easily accessible and cost-efficient, many pure play e-commerce actors choose a parcel carrier to bridge the last mile. But the person making the delivery is generally not able to provide the shopper with extra information or value added services. So, this leads to a reduced delivery service towards the customer. And the picking is one or more dedicated distribution centres. - Click and mortar: The companies, which use this model, dedicate to sell their products by e-commerce and by physical shop. Inside this model, the main delivery model is: o Van delivery: 32 David Bardasco San José 1.3.3.3. Problems in Last Mile Logistic In line with Brown and Guiffrida (2014), the last mile problem comprises one of the most costly and highest polluting segments of the supply chain in which companies deliver goods to end costumers. But Frazer (2000) already identified that time constraints, poor quality of home delivery services, and lack of variety of delivery options to be the influential factors that make home delivery the weakest link in the Internet chain. Charantan (2001) reports that non-satisfactory delivery schedule topped the list (34%) of dissatisfaction with e-commerce. It means the importance of delivery for online shopping to the customers. Specifically, the 40% of these prefer easier delivery or collection in addition to cheaper offers and lower delivery charges offered by e-tailers. And a survey of 100 companies conducted by Consignia (2001) revelled that 58% of the respondents ranked delivery at the time and place that is convenient to the consumer as the second most important factor influencing the market. Aized and Srai (2014) say that the last mile is considered one of the most expensive parts of the supply chain and accounts for 13% up to 75% of the total supply chain costs. Specifically, next to the picking and packing operations, home delivery is the major cost driver in online grocery shopping (Mikko Punakivi and Kari Tanskanen, 2002). To confirm this affirmation, an empirical study (Ring and Tigert, 2001) examining delivery models adopted by grocery retailers in the USA, UK and Europe found that the two killer costs facing pure Internet grocers are the picking costs and the delivery costs. All these high costs provide an opportunity for companies to achieve substantial efficiencies through optimal planning and proper execution of a delivery plan which may involve analyses to redesign the overall distribution network, establishing more efficient routings, changing delivery zonings, or upgrading to a more fuel-efficient transportation fleet (Brown and Guiffrida, 2014). Crowdsourcing Logistics in B2C e-Commerce 33 Managing this portion of the supply chain (last mile) has been a particular problem from logistics infrastructure standpoint, most notably because of tradeoffs between routing efficiency and customer convenience. And to select the method which consumers place orders can have a significant impact on transaction costs and customers service (Kull et al., 2007). So, one of the most difficult problems for logistics management is such a fulfilment process is to solve how to plan and implement a cost effective delivery operation, in a dynamic environment, where commitments to customers must be given while orders are being received (Slater, 2002). In practice however, for many B2C companies, as suggested by Newton (2001), the cost savings promised by ecommerce are eaten up by high delivery expenses. · Problems: Some of the analysed papers, specifically 33% (5 papers), talk about the main problems in the Last Mile Logistic and identify them. Basically, these publications define two types of problems in the delivery at home, which are: - High degree of failed deliveries (Weltevreden (2008); Xu et al. (2008); Edwards et al. (2010); Aized and Srai (2014)): The incidence of first time delivery is one of the most common problems in the Last Mile Logistic and it is obvious that it causes important extra costs for the companies. Such as say Xu et al. (2008), the most traditional delivery option used by many retailers for home shopping is using courier and postal services. Ant the delivery time varied including same-day, next day and multi day delivery. It should be noted to that many parcels do not fit through mail or letterboxes or require consignee signature, which implies that customers need to be at home when the parcel is delivered (Weltevreden, 2008). The problem is that the most common for people is not to be at home during the working day. So, they are not at home when the most home 34 David Bardasco San José deliveries are made (Edwards et al. 2010). Talking about some dates, the working households increased by 22% between 1992 and 2006. It means, the incidence of failed deliveries has increased and nowadays follows the same direction, because every time people use more ecommerce and it increases home deliveries. A survey shows that 34% of the respondents indicated that the best delivery time slots from them would be between 6 pm and 8 pm (Xu et al., 2008). But this isn’t a good option, because it creates large demands of delivery for a short busy period, which results in the delivery fleet runs at low capacity for 80% of the day, then at fall capacity for the sent. And this problem is called “Not at home” problem, because the people is not at home at home time delivery (Xu et al., 2008). And it is considered one of the most critical factors for the success of the home delivery operations, because it causes higher operating costs for retailers and carriers, as result that need to be redelivered or returned the packages to the sender and then repeat another time all the delivery process. And also, inconveniences to customers that lead to lower satisfaction, because they don’t have their purchase when they want. So, the broken promises and unmet expectations of last mile e-commerce left both consumers and investors dissatisfied. For these reasons, it is working to find alternatives solutions to try to resolve this problem. - The return of unattended goods (Edwards et al. (2010); Aized and Srai (2014); Hbner et al. (2016)): The incidence of return of unattended goods is other of the most common problems in the Last Mile Logistic and it is obvious that it causes important extra costs for the companies. Customers return items for a number of different reasons. Typically between 25 and 30% of all non-food goods bought online are returned compared with just 6-10% of goods purchased by traditional shopping methods (Edwards et al., 2010). Crowdsourcing Logistics in B2C e-Commerce 35 According to Hbner et al. (2016), one of the drawbacks of online shopping is the customer’s inability to see and feel the product before purchasing it. Especially in online grocery this becomes a common factor as consumers have general reservations about the retailer selecting and touching their food and consumer about the quality. About how resolve this problem, it depends of the type of company. It means, in grocery stores the retailers can offer to customers a moneyback guarantee, check and return at reception, return by CEP delivery, or acceptance and refunding (Hbner et al. (2016)). And in other industries, the typical return channels are: return items to a physical store or send items back through the standard postal service. And between these channels, approximately half of returns are via carrier collection and half by post (Edwards et al, 2010). It should be noted to that some delivery companies offer an other channel, and it consists to send vans on separate pick-up runs dedicated solely to collecting returned items. It is important to say that there are some companies that allow choose the customers, which will be the return channel, that use. But it only makes more complex the situation, because the retailers have to manage more services. And it is obvious that all these generate important extra costs for the retailers. For these reasons, it is working to find alternatives solutions to try to resolve this problem. It should be noted that apart of these that are the main problems, Aized and Srai (2014) talk about two more, which are: High degree of empty running of vehicles and low volume of delivery goods. 1.3.3.4. Alternatives Models to Last Mile Logistic Then of analysed all the publications, some alternatives models to Last Mile Logistic have been identified with the objective to solve the different problems in 36 David Bardasco San José Last Mile and to try to offer more possibilities to the final customer. Basically, two alternatives are: · Home delivery: This concept is focused in that goods are delivered to the store and customers perform the picking and final delivery to their home themselves. So, home delivery concept provides additional customer satisfaction. The direct concept offers consumers two models: an attended model of reception and unattended model of reception. - Attended Delivery (Weltevreden, 2008; Al-Nawayseh et al., 2013; Hbner et al., 2016): The 20% (3 papers) of analysed papers talk about the attended delivery. Figure 2: Diagram of Alternatives Crowdsourcing Logistics in B2C e-Commerce 37 According to Al-Nawayseh et al. (2013), the attended delivery is the attended home reception where customers usually choose the delivery place and time window to receive their delivery within it. So, this model implies that the customer has to be at the point of reception within the time window that she/he has selected to accept the delivery and it implies that she/he is waiting for her/his delivery. On the other hand, one of the retailers’ objectives is maximizing vehicle utilization and minimizes transportation costs to get a certain level of customers’ service and the satisfaction of these. And it requires dynamically assigning delivery time slots as new orders arrive and the dynamically creating and adjusting delivery routes (Hbner et al., 2016). It means the vehicle routing becomes more complex, because it has to due to customers’ time restrictions and it is obvious that it generates costs. Analysing the problems about this delivery model, basically there are two big problems, which are: o The high demand on certain windows might complicate the service, because is possible that the provider cannot make all the deliveries at the same time, so it generates capacity problems. o The problem of failed delivery: It has been told in the previous part. It means that the customer is not available for order the reception when she/he had said, and as result the truck returns without fulfilment and all the problems that it implies. In spite of this complexity that this model generates for all participants, the attended delivery model is used for home delivery of grocery goods across Europe regardless of market proliferation 38 David Bardasco San José (Hbner et al., 2016). Specifically, in most countries, this model accounts for the largest share of last mile deliveries. One alternative practice of attended delivery is the Service Posts, which following is told:  Only one of the analysed papers (Weltevreden, 2008) talks about this alternative practice to last mile delivery. This alternative consists in that the parcels are delivered to a store, petrol station or post office where customers can pay, collect and return their parcel. Specifically, Weltevreden (2008) defines it as a shop in shop concept. It should be noted that at a service point, the persons who manage the collection procedure are the shop assistants. This practice offers some opportunities, which can be positive for all the parts. The Service Points are often located in a store and it means that is combined the collection of the parcels with other shopping activities. So, it can offer retailers prospects of a revenue increase. And it is really good too for the delivery companies, because this way they can combine the delivery of parcels with the regular supply of the shops. - Unattended Delivery (Kämäräinen and Punakivi, 2004; Xu et al., 2008; Hbner et al., 2016): The 20% of analysed papers talk about the unattended delivery. In line with Xu et al. (2008), the original concept of unattended delivery is simply leaving an item on someone’s doorstep, or in the garden shed. So, in this model retailers can deliver online purchase of whether the customer is at home or not, because the shopping basket Crowdsourcing Logistics in B2C e-Commerce 39 is place in front of the customer’s home to be collected when she/he arrives. Kämäräinen and Punakivi (2004) with their simulations suggest that this delivery model reduce home delivery costs considerably, specifically by up to 60%. But the problem is that this hasn’t been widely used, because it requires investments and commitment from the customer. For the other hand, Hbner et al. (2016) say that the cost of delivery can be reduced by up to 40% compared to attend home delivery with a reception box for unattended home delivery. The main problem of this is that it brings many security concerns particularly when items are perishable or have high value. But it is a really good option to solve the problems of home deliveries fail, because in 50-60% of households no on is at home during the normal workday an average of 12% of home deliveries fail (Hbner et al., 2016). This delivery model has different advantages. Talking about the logistics, this eliminates tight time slots and capacity problems resulting from uneven demand during working hours. It means that demand peaks are evened out. And unattended reception shortens the working hours for the distributor. So, it eliminates the redelivery cost when the customers are not at home at their selected delivery time slot. It is important to say to that normally retailers that follow this model will charge additional fees if the customer is not able to receive their delivery in the agreed time slot. The most common alternative solutions for unattended reception are:  Delivery Boxes: This practice is focused on the use of insulated box containing the goods is delivered to the customer and attached securely in a 40 David Bardasco San José locking device bolted on the building wall. Then the empty boxes are collected on the day following delivery or later. So, basically the delivery boxes are insulated boxes with a docking mechanism that are returned to the retailer. And there are 2 ways to pick up empty box:  Pick up of the delivery boxes is done at the sect delivery.  Pick up of the boxes is done separately on the day after delivery. Regarding to Punakivi et al. (2001), this solution enables a faster growth rat and higher flexibility of the investments, because of a smaller investment required per customer. The drawback is the additional cost of collecting the empty boxes.  Reception boxes (Locker Point): This solution consists in a collection of lockers. Specifically, the parcels are delivered to the lockers point where customers can pay, collect and if necessary, return their parcel. It should be noted that the locker points employ baggage lockers technology and use PIN codes to control the delivery by the carrier and the collection of the parcel by the customer (Weltevreden, 2008). And usually these boxes are installed in the consumer’s home yard or garage. In line with Kämäräinen and Punakivi (2004) and from point of view in B2C environment, the reception boxes have some advantages, which are: time savings, flexibility, independence of the supplier timetable, no need to carry purchases, decreasing of impulse buying and systematic purchasing. But there are also some disadvantages, which basically are: high investment and limited access to deliver to the boxes in B2C. Specifically, high investment is considered the major obstacle, because the box can Crowdsourcing Logistics in B2C e-Commerce 41 cost between 1000 and 2000 euros depending on the type of box (Kämäräinen and Punakivi, 2004). It is important say that the reception box is widely seen as a potential home appliance of the future and there are already many manufacturers in the market. However, utilization rate of the reception box is usually poor (Kämäräinen and Punakivi, 2004).  Shared reception boxes: This solution is focused in usage of shared reception boxes also known as CDP. About the characteristics of the shared reception box unites, these have various amounts of separate locker, and the separate lockers have electronic looks with a changing opening code to enable shared usage of the lockers using a mobile phone (Kämäräinen and Punakivi, 2004). As previously it said, this concept is also known as the automated collection and Delivery Point (CDP). And at the CDP the ordered goods can be stored until the customer is able to collect the delivery. In line with Mikko Punakivi and Kari Tanskanen (2002), using this concept, the utilisation rate of the facilities would be higher than in the case of customer-specific concepts. The main problem is the high investment involved in unattended reception facilities could be solve by sharing the responsibility. But there are others factors that affect home delivery transportation costs in this model, which are (Mikko Punakivi and Kari Tanskanen, 2002): o The capacity of the shared reception box unit, that is, the number of separate lockers. o The number of separate reception box units. o The utilization rate of the shared reception box units. 48 David Bardasco San José the own web of the project and other sources of information of Internet. All with the objective of to check and to claim that the identified cases know suitably the determined requirements. The projects that have been found are detailed below: No TITLE 1 AMAZON FLEX 2 DELIV 3 DOORDASH 4 INSTACART 5 MyWays 6 POSTMATES 7 RICKSHAW 8 SIDECAR DELIVERIES 9 UberEATS 10 UberRUSH 11 ZIPMENTS Table 10: Projects 2.2. CLASSIFITACION AND SUMMARY OF THE PROJECTS One time that the projects have been identified, it has proceeded to analyse them. In order to analyse them of the best possible way and to study them suitably, it has carried through different classifications. This form, it can see clearly different key aspects about the projects and can draw convenient conclusions. The classifications are: 1) Basic: It is focused in to determine the basic information of every project that mainly it is focused in two aspects: the project and the company that carry through the project. 2) Of the Crowd: It is based on to define the main characteristics of the Crowd that must have to work in the pertinent project. Crowdsourcing Logistics in B2C e-Commerce 49 3) Deliveries: It is focused in the key aspects in the deliveries which basically are: the pick up, the delivery areas and what the delivery is. Is important to say that the analysis is carried out of a total of eleven projects. 2.2.1. Basic Before to process to do the basic classification is very important to know what the companies think about their projects about Crowdsourcing Logistic. So, next it can see how the companies define their projects: - Amazon defines Amazon Flex, the project about the last mile delivery as collaborative economy in the logistic where the independent drivers deliver parcels. - Deliv defines its project as a same day delivery service revolutionizing how online and in store customers get their goods. - DoorDash defines its project as a technology platform that connects local business to people. They aim to make every city smaller by bringing the food for people-faster, fresher, and from farther away. - Instacart defines its project as a grocery delivery service that delivers in as little as an hour. - Deutsche Post DHL defines MyWays as a completely new delivery service for people who need parcels delivered where they want, when they want. - Postmates defines its project as a revolutionary urban logistic and ondemand delivery platform that connects customers with local couriers. - Rickshaw defines its project as a same day delivery platform that allows any business to schedule deliveries to their customers without the hassle of managing a fleet of cars and drivers. 50 David Bardasco San José - Sidecar Deliveries defines its project as a same day service for local business whereby goods, food and flowers were to be delivered to local consumers using its existing pool of drivers. - Uber defines UberEats as delivers the best meals from favourite local restaurants in 10 minutes or less. It brings customers a meal on demand with none of the hassle. - Uber defines UberRush as a connection between customers and couriers to make a delivery. Customers can use it to power faster deliveries and returns. - Deliv defines Zipments as a community base logistics platform providing business and individuals with the fastest, most affordable same day delivery service available. It is important to say that actually there are two projects that aren’t working which are MyWays and Sidecar deliveries. Next it can observe the basic classification of the studied projects: Crowdsourcing Logistics in B2C e-Commerce 51 Table 8: Basic Classification of the Projects 52 David Bardasco San José Regarding to the basic datum of the projects, to emphasize that the origin of these projects has place between the year 2011 and 2015. Being the year 2014 when fewer projects carried out, specifically one. However, during the years 2013 and 2015 were when more projects did (3 every year). As regards at the companies which come from the project, they have been analysed by the type of company that they are in this moment. Specifically, four of these that is a 36% are multinationals companies and the rest are start-ups. And in relation with this aspect, it can observe that the project name and the company name is the same in the case of start-ups. Too is important to say that as well as much Deliv as Uber has two projects, but Deliv bought one of these of a start-up, specifically of Zipments which nowadays still is called of the same name although it is property of Deliv. Related to the origin country of the companies, ten of these (91%) are from America, specifically from United States of America and only one (9%) is European, specifically from Germany. Being more specific, inside in American companies, the 64% has origin in San Francisco, California. On other hand, it has identified if the activity of the project is a service or a network. It means that it has considered a network if the project take part of the logistic network in the own company (the company has her products, her warehouses, etc.). And it has considered a service if the project isn’t involved in a network, so the activity is only to satisfy a customers’ necessity and is the main and unique activity of the company. Then of to analysed this aspect, it can see that more or less the 80% of the projects are service and the rest (20%) are networks following the consideration. 2.2.2. Crowd Then of to identify, analyse and classify the Crowd about the different selected projects, it can observe that there are series of requirements to can take part of this. These requirements is focused mainly in: - Be at least the minimum age. Crowdsourcing Logistics in B2C e-Commerce 53 - Have the type of required and permitted vehicle to carry out the pick up delivery and the deliveries. - Have a phone with determined characteristics. - Have a series of documents with the requested conditions. - Be able to have the physical capacity to lift packages of a certain quantity of weight. It should be noted that to carry through the logistic job (pick up and deliver the customers’ purchase) the member of the Crowd receives a reward. Next it can observe the crowd classification of the studied projects: 54 David Bardasco San José Table 9: Crowd Classification of the Projects Crowdsourcing Logistics in B2C e-Commerce 55 First of all to emphasize that the range of minimum age to can be member of the Crowd and to make the requested jobs is between 18 and 21 years old. Specifically of the analysed projects, the 45% demand to be at least 18 years old, the 18% demand to be at least 19 years old and the rest demand to be 21 years old. As regards how the pickups and the deliveries are carried out to the customers basically it can distinguish between motor vehicles (car, van, motorcycle), bike or on foot. The 45% of the projects only allow making their job by motor vehicles, where the car is with difference the most requested. It is important to say that the 55% of the analysed projects allow the possibility of not to use a motor vehicle, it means that their offer the alternative to make the pickups and deliveries by bike or walking. Specifically of this 55%, the 50% of the projects offer the alternative of on foot or go by bike and the other 50% only offer the alternative of go by bike. Anyway there are some projects as for example Amazon Flex that wants to offer opportunities to deliver via bike or on foot in the future. It should be noted that the vehicle must be property of the member of the Crowd. There are some projects that request some extra requirements as for example, the project of Deliv request to their members of the Crowd that their cars must be of year 2000 or newer and with air conditioning. To establish communication between the company and the member of the Crowd that is for the member knows where must go to pick up the delivery and where he must deliver it, all the projects use the same medium to receive the orders, the phone. The 91% of the projects allow have an operative system Android or iPhone, anyone of these is valid. However, the unique project that only allows one operative system (Android) is Amazon Flex. With respect to an other interesting datum about this aspect, it should be pointed out that some projects, as for example Instacart demand to the members of the Crowd that have a recent smartphone, it means as minimum a iPhone 4 or Android 4. With the objective to guarantee a right running and to offer security to their customers, all the analysed projects request a series of documents to the 56 David Bardasco San José members of the Crowd which basically they are: have a valid driver’s license, an insurance of the vehicle and pass a background check. Some projects request more requirements about this aspect as for example the project of DoorDash which requests at least 2 years of driving experience to their members or UberRush and UberEATS which request at least 1 year. Some analysed projects request that the members of the Crowd have some physical abilities which they are linked to be able to lift weight. All this with the objective of to move correctly the customers’ delivery. Specifically, the 45% of the projects don’t request a minimum (or don’t think that it is a key aspect) and the rest (55%) consider that it is important. Of the projects that think that this ability is important which are: Deliv, Instacart, Rickshaw, UberEATS, UberRush and Zipments, they consider basically that the couriers that deliver by bike or on foot must be able to lift 30 pounds and the couriers that deliver by motor vehicle must be able to lift 50 pounds. As regards to the availability of the members of the Crowd is obvious that it always goes in function of the delivery volume that could be and the availability. But all the projects offer big flexibility to their members to choose their work hours. Some of the projects offer blocks of time to work, as for example Amazon Flex which allow choose any available 2,4 and 8 hours block of time to work the same day. Or the project of Rickshaw that allow choose any available 4 and 8 hours block of time to work the same day. In some of the analysed projects, they request to have a customer service skills to their members of the Crowd. Specifically these projects are: Deliv, Instacart, Rickshaw, Sidecar Deliveries and Zipments. Also it is important to emphasise that some of the selection process to be member of the Crowd are more demanding than others. It means that in some of these they only request to fulfil the main characteristics that are requested. However, in others as for example in the Deliv project, it carries out an extensive filtering process which includes interviews by video. Or for example, Postmates that does to pass a deliveries test and a personal interview. Crowdsourcing Logistics in B2C e-Commerce 57 Finally, with respect to the reward for the completed jobs, all the analysed projects determine the price of the member of the Crowd per hour and not for parcel. The compensation range is between 18$ and 30$ per hour. But the most common is that company pays up to 25$ per hour (55%). Only there are two projects (UberEATS and UberRush) where the company could pays more (up to 30$ per hour) and they are of the same company, Uber. It is important to emphasise the case of MyWays which allows at the customer to decide how much will pay for the service. Then the company take a part of this quantity, specifically a 10% and the rest is for the member of the Crowd. Other aspect important about MyWays in difference with the other projects is that pays in Kronas, because it carries out in Stockholm (Sweden) and the other projects pays with dollars. 2.2.3. Deliveries Next it can observe the deliveries classification of the studied projects: 64 David Bardasco San José from the Local Shop to every customer and of the weight of the delivery, because every transport has a capacity. ·Cost: The model says that is the total cost of the logistical service (Crowdsourcing Logistics) for the company in function of the characteristics of the Member of the Crowd. And basically it shows the main costs of this, which are: Unit Cost of Acquisition (Pick up Cost and Deliveries Cost), Information and Control Cost, Cost of Launch of the service and Breakage Cost. · Emissions: The model says that is the total emissions of the logistical service (Crowdsourcing Logistics) with the chosen transport and the determined routes. 3.2. Context It should be noted that to carry out the creation of the model is necessary to answer and fix some aspects that are basic in the previously cited practice. These are: · Who buy the service? Then to analyse different projects about Crowdsourcing Logistics, it has been observed that almost cases the service was bought by Local Shops. So, in the proposed Model, the Local Shops will buy the service, because nowadays the projects are focused in this system. About the Local Shops and in relation with the topic of this project (Crowdsourcing Logistics in B2C), it has been considered in the model that the Local Shops used to sell the traditional method and e-commerce. It should be noted that with the e-commerce and with this new logistic service the final customer could receive his purchases at home without have to move of his house. · Who is the service provider? Crowdsourcing Logistics in B2C e-Commerce 65 Then to analyse different projects about Crowdsourcing Logistics, it has been observed that almost cases the service provider are start-ups. Specifically, the start-ups are born to satisfy the customers’ logistic necessities. So, in the proposed Model, the service provider is a start-up, because nowadays the most projects come from this type of company. · Who makes the delivery? The delivery will make by the Crowd. The Crowd is a number of people that want to do this logistic service, offer their skills for this and have the characteristics and the requirements that are necessary for every company. Then to analyse different projects about Crowdsourcing Logistics, it has decided that in the proposed Model the Members of the Crowd have the most common characteristics and skills of all the studied projects. Specifically, these are: · Type of transport to make the deliveries: Car, Motorbike, Bicycle or on Foot. · If the Members of the Crowd make the deliveries with some vehicle, they have to have their own vehicle. · Be at least 18 years old. · Have a phone. It could be Android or iPhone. · Have a valid driver’s license, insurance and pass a background check. · Professionalism and good communication skills when dealing with clients. · Ability to lift packages that range from 10-25 kg in and out of vehicles, up and down flights of stairs. 66 David Bardasco San José 3.3. Scope The scope of the model is delimited basically by the hypothesis establishes, which are: 1) It will be assigned one Local Shop, which will have different deliveries to make to its customers, at one Member of the Crowd. Specifically, it will be assigned the nearest at the Member's of the Crowd location. 2) It is considered that the Member of the Crowd must complete all the deliveries of the Local Shop that has been assigned. 3) There isn't a priority order in the deliveries in the Local Shop. Specifically, the priority order will go in function of the optimal route. To calculate the optimal route has been used Clarke's and Wright's algorithm which is traditional for VRP (Vehicle routing problem) and is used to plan the routes. It has considered convenient to use this algorithm, because in this type of service the done distance is fundamental, basic factor. For this reason, it is very important to select the best route to optimize costs and emissions. 4) It is considered that the demand of the service by the Local Shops is produced all at the same time (in the same hour). For this reason, every Member of the Crowd has to go to one specific Local Shop, because the first delivery of every Local Shop is more or less at the same time. 5) The demand tax of the service of the Shops must be less or equal than the number of the Members of the Crowd, because if not it won't be possible to provide all the Shops that need the services. 6) The Local Shops and the final customers are in the urban city. 7) The distance is considered in Euclidean distance with the coordinates in Km. Crowdsourcing Logistics in B2C e-Commerce 67 8) The Member of the Crowd can deliver by car, by motorbike, by bicycle or on foot. 3.4. Methodology The methodology applied to this model is the next: - First of all, the required data are entered. - Then, the optimal route is calculated with the attached conditions. And the algorithm use for this is the Vehicle Routing Problem, specifically Clarke’s and Wright’s algorithm. The resolution of this algorithm (parallel version) is:  First Step: It is calculated all savings aij with respect to the deposit for all the customers’ pairs “I” and “j”.  Second Step: To order all the customers' pairs of the biggest savings to the smaller savings.  Third Step: While there are customers to assign: o Select the pair (i, j) that isn't assigned with minor or equal demand at the highest capacity of the route and with the biggest savings. Cases: I) If i and j aren't assigned: To open a new route with the extremes i and j. II) If i is assigned with the extreme of an open route R with a bigger o equal capacity than the demand of j: To assign j a R. III) If j is assigned with the extreme of an open route R with a bigger o equal capacity than the demand of i: To assign i a R. 68 David Bardasco San José IV) If i is the extreme of an open route R and j is the extreme of an open route R', and the sum of the loads of R and R' don’t exceed the biggest capacity of the route, merge R and R'. - One time that the route is calculated, next is applied the costs and environmental model with the formulas, the datum and variables, and the restrictions corresponding (in the next section are told all these concepts). 3.5. Cost Model Next the cost model is told with his formulas, datum and variables, and the restrictions. 3.5.1. Formulas The costs that have been considered basically are: · Unit Cost of Acquisition (Cu): This is the cost for the service of one member of the Crowd to one local shop with all their deliveries that the shop has in that moment. · Cost of Launch of the service (CL): This is the cost that implies to launch the service every time that one local shop needs, requires the service of Crowdsourcing Logistics (management, personal, etc.) · Breakage Cost (CB): This is the cost that implies not to fulfil with the demand of the Local Shops, because there aren't enough Members of the Crowd. So, the formula to calculate the delivery cost is: Units: (€/delivery) · Unit Cost of Acquisition (Cu): The formula is the next: Crowdsourcing Logistics in B2C e-Commerce 69 Units: €/delivery - Cpick_up: This is the cost that implies the Member of the Crowd goes from where she/he is to the nearest Local Shop. o Cpick_up:  Cpick_up_Transport: This is the cost that implies the transport of the Member of the Crowd to the nearest Local Shop.  Cpick_up_Personal: This is the cost that implies the required, used time for the Member of the Crowd.  ES: This is the number of deliveries that has to make the local shop that has been selected. - Cdeliveries: This is the cost that implies the Member of the Crowd goes from the Local Shop to make all the deliveries that the shop has. Also it includes the time that the Member of the Crowd needs to pick up the deliveries in the shop and the time that she/he needs to deliver every delivery to the customer with the correct service. o Cdeliveries:  Cdelivery_Transport: This is the cost that implies the transport from the Local Shop to all the deliveries that the shop has to do.  Cdelivery_Personal: This is the cost that implies the required, used time for the Member of the Crowd.  ES: This is the number of deliveries that has to make the local shop which has been selected. 70 David Bardasco San José - CFailed_deliveries: This is the cost that implies to make failed deliveries. It means that the final customer isn't at home, so the MC has to come back to the Local Shop to leave the delivery. o CFailed_deliveries:  CFailed_delivery_Transport: This is the cost that implies the transport from the Local Shop to all the failed deliveries that the shop has to do.  CFailed_delivery_Personal: This is the cost that implies the required, used time for the Member of the Crowd.  ES: This is the number of deliveries that has to make the local shop that has been selected.  YFail: · Cost of Launch of the service (CL): The formula is the next: Units: €/delivery - CLaunch: It is the cost to launch the Crowdsourcing Logistics (management, team work, communication, data transfer, etc.). - ST_Deliveries: It is the number of shops that are partners of the company and need to make deliveries. · Breakage Cost (CB): The formula is the next: Crowdsourcing Logistics in B2C e-Commerce 71 Units: €/delivery - ST_Deliveries: It is the number of shops that are partners of the company and need to make deliveries. - NMC: It is the number of the members of the Crowd. - CBreak: It is the cost to launch the Crowdsourcing Logistics (management, team work, communication, data transfer, etc.). - YB: 3.5.2. Explanation of datum and variables The datum and variables that have been considerer in the previous formulas are: 72 David Bardasco San José 3.5.3. Restrictions The restrictions and conditions that have been considerer in these formulas are: 1) Every Member of the Crowd only can make the delivers with one type of transport. It means by car, by motorbike, by bicycle or on foot. So: 2) About the fuel, it is only necessary in motor vehicles. Specifically, the car is the unique transport that can work with petrol or gasoil, but only with one of this. And the motorbike only works with petrol. So: Crowdsourcing Logistics in B2C e-Commerce 73 3) About the amortization and use, if the Member of the Crowd makes the delivery on Foot, it hasn't any cost of amortization and use, because she/he doesn't use any machine. So: 4) In the cases that have been necessary to calculate the transport time, it has used this formula: 3.6. Environmental Model Next the environmental model is told with his formulas, datum and variables, and the restrictions. 3.6.1. Formulas The emissions that have been considerer basically are: · Pick Up Emissions (EPick_Up): These are the emissions that are generated, because the Member of the Crowd goes from where she/he is to the nearest Local Shop with her/his vehicle. · Deliveries Emissions (EDeliveries): These are the emissions that are generated, because the Member of the Crowd goes from the Local Shop to the different places where are the deliveries with her/his vehicle. · Failed Deliveries Emissions (EFailed_Deliveries): These are the emissions that are generated, because the Member of the Crowd can't make the delivery (the customer wasn't at home). 80 David Bardasco San José To carry out he related simulations has set out a baseline scenario, which it has been changing variables in order to achieve the greatest possible number of data with the different possibilities and can to draw better conclusions. Specifically, for the above scenario the following characteristics has been considered: - The set ranges of the main variables have been: o Number of Members of the Crowd: 2,4,6,8 and 10. o Number of shops with deliveries to make: 2,4,6,8 and 10. o Number of customers of the most nearly shop from the selected member of the Crowd: 1,2,3,4 and 5. - The block work of each member of the Crowd is approximately 2-4 hours. - Simulations have been made for transport 4 considered, which are: car, motorbike, bicycle and on foot. And given that the car has the variant: diesel or petrol. - To fix the coordinates of each variable that require location, first of all has been considered a delivery area of 5 x 5 km, i.e. 25 km2. All with the objective that the obtainment of datum was the most realistic and coherent as would be possible. Also, it is important to say that the coordinates have been given a factor of 0.9713, because it has been considered that using Euclidean coordinates was necessary to use a factor, because the streets do not follow that way and sometimes there that deviate from the path. - It has been considered coordinates for the location of the Member of the Crowd, for the location of the shops and for the location of the customers. And all these has been made completely random and through of function of Excel. And the number of failed deliveries too. Crowdsourcing Logistics in B2C e-Commerce 81 - It has been considered that each transport could bting a certain amount of packages. Specifically, the amounts set has been as follows: o Car (whether or not diesel or petrol): 5 deliveries. o Motorcycle: 3 deliveries. o Bicycle: 2 deliveries. o Walk: 2 deliveries. Obviously, this fact has significantly affected the distances and the route has made the member of the Crowd, because if the transport cannot cover all the deliveries, it means that has to come back to the shop and then go to the other customers. The basic scenario datum, which has made through Excel, has been the next: Figure 6: Basic Scenario 82 David Bardasco San José And from there, it has raised the following changes of variables to study better the model and thus make a more profound and truthful analysis. Follow, it can see an abstract of all the cases simulated with the different variables: In each case those shown has set a number of members of the Crowd, a number of shops that had deliveries and a number of customers in a particular store, which was the one that was closer regarding the location of the member of the crowd. It is important to say that only are analysed deliveries perform a single store by a single member of the Crowd, but it is necessary to consider all the facts, because this affects and have an effect on the final price and emissions of each delivery. Figure 7: Cases Crowdsourcing Logistics in B2C e-Commerce 83 4.2. Costs: Summary and conclusions Following the simulation procedure cited in the preceding paragraph and with the corresponding model, there have been a total of 25 simulations in relation with the costs that generate every delivery (with fixed data and hypothesis considered). Each of them in turn, with their respective variants (car works with petrol, car works with diesel, motorcycle, bicycle and walking). In the section Annexes (Results of the simulation), it can see all the results obtained in each case around costs. After analysing the results, then are emphasised those aspects around costs that are considered most important and relevant to them into account, as well as to draw conclusions about the practice studied (Crowdsourcing Logistics) and the proposed model. And taking into account the characteristics and fixed assumptions, hypothesis. Costs mainly depend on the distance travelled and the time spent by the member of the Crowd as well as the type of transport. Analysing the results, it can be seen that by performing deliveries by bicycle or on foot saves on consumption and amortization, but instead in terms of time costs increase. So that the most balanced option is that the member of the Crowd makes the deliveries by motorcycle, since it is the most economical and therefore the most interesting option for businesses and for the realization of Crowdsourcing Logistics. The cost of delivery is in a range between 2,43 and 38,62 Euros per delivery. Emphasize that much of the cost is in the part of customer’s delivery of his purchase, as well as the breaks in service when there are these. As regards the other costs involved in the proposed model, they are practically negligible. 84 David Bardasco San José Note that from the perspective of costs, the break of service, i.e. not having sufficient resources (members of the Crowd) generates significant implications in costs and dimensions lot these. In the model created, it has sought to highlight this aspect, for this reason it has been given considerable cost to the fact that has breakages, because it generates customer dissatisfaction and because it is one of the main problems that the practice cited too. This problem cited basically focuses on that if the number of members of the Crowd is less than the number of stores that have to make deliveries, it can not fulfil all the demand and therefore, it doesn’t allow to serve to customers, which obviously it is negative. So, from the perspective of costs both the number of members of the Crowd as the number of stores that have deliveries to make, are very important, because they are directly involved with the cost of breakage, which can increase lot delivery costs. Also, it must say that the bicycle option is a very tempting option and which through simulations has been seen that can play a very important role in this practice, since the costs are very close to the motorcycle, and taking into account which is a clean transport, i.e. not contaminated (the emissions are analysed in the next section). Therefore, it can be a very good alternative to offer Crowdsourcing Logistics with transport by bicycle to the members of the Crowd. But instead it requires more time and the capacity in some cases is a problem, because bicycle can not bring the same deliveries than the car. So in this aspect, is far of this. Next it can observe the case 1 at the level of costs, where it can see the option of walking in this practice still far away, since this type of delivery is very expensive due to the time required. In this graph can also appreciate the strong threat of bicycles as a clear competitor to the motorcycle, leaving in third place the car, which loses strength. The car is not the most suitable transport for this service, because in the city is hard to go by car. Anyway it remains an important alternative, because it allows lots of deliveries on the same route, since the car is the transport of more capacity that has been set for this practice. In the graph it can see that increasing the number of customers, the delivery costs drop, Crowdsourcing Logistics in B2C e-Commerce 85 which generates that have to squeeze the maximum. All this with the finality of to return the fewest times possible to the store. With respect to the graph of the case 25, note that follows the same line as above, but with some minor differences, which are linked to the distance. In conclusion, it can say that Crowdsourcing Logistics from the perspective of costs is focused on the distance travelled, in the time required and the transport. Not forgetting, breaking service that apart from generating customer dissatisfaction can generate very significant increases in costs. Therefore, after performing simulations, it can say that from this perspective, the model is consistent, since Figure 8: Costs Case 1 Figure 9: Costs Case 25 86 David Bardasco San José the results are very similar to the reality (with the limitations of the model). And in turn, it can open the doors to the viability of Crowdsourcing Logistics, because delivery cost is quite similar to the current main competitors and main alternatives. 4.3. Emissions: Summary and conclusions Following the simulation procedure cited in the preceding paragraph and with the corresponding model, there have been a total of 25 simulations in relation with the emissions that generate every delivery (with fixed data and hypothesis considered). Each of them in turn, with their respective variants (car works with petrol, car works with diesel, motorcycle, bicycle and walking). In the section Annexes (Results of the simulation), it can see all the results obtained in each case around emissions. After analysing the results, then are emphasised those aspects around emissions that are considered most important and relevant to them into account, as well as to draw conclusions about the practice studied (Crowdsourcing Logistics) and the proposed model. And taking into account the characteristics and fixed assumptions, hypothesis. Emissions depend mainly on the distance travelled and of the transport used. Obviously, neither transport by bicycle or walk transportation generate emissions. Therefore, this practice has great potential to be friendly with the environment and does not generate emissions, i.e. not to pollute. So that it can be considered a green practice, because when the members of the Crowd use these types of transport, do not pollute. However, it should be noted that members of the Crowd must travel more large distance and having to use their own body, it takes longer and fatigue by members of the Crowd. Note that in the rest of conclusions that are drawn around emissions has not been talked about bicycle and walk, since it has been considered evident that these do not generate any emissions and therefore, this study is not necessary (in this aspect), because they are the best option (talking about emissions). Crowdsourcing Logistics in B2C e-Commerce 87 When the member of the Crowd must make deliveries to more than three customers, the transport that generates fewer emissions, i.e., the least polluting, is the car, specifically that works with diesel. And this is mainly due to the capacity, because the car can load more deliveries unlike the motorcycle. However, if it must make less than two deliveries and therefore, the member of the Crowd makes the direct transport, i.e. goes from the shop to customers without having to go back to the shop, it is considered that motorcycle is the most appropriate transport (in relation with emissions), generating fewer emissions than a car. It should be noted that the fact that the number of stores that have deliveries to make or the number of members of the Crowd, has been seen that is not a key factor or that involving changes in the amount of emissions per delivery. Emissions range are from a value range of 0 to 1,40 KgC02 / delivery. Note that much of the emissions generated are in the part of the delivery of the purchase at the end customer, always that has performed more than one delivery, because it is when the member of the Crowd must travel more distance. When more deliveries are made, emissions generated by these deliveries are spread and is much more significant than a single delivery. Following is attached a chart of Case 1, which it can see the emissions generated by the car with petrol, diesel and motorcycle. And it has been changed the number of customers. Also, it can observe that when it just has to make a single delivery, that is, a single customer, emissions are higher than when have to make some deliveries. Therefore, when it must carry out more deliveries, emissions spread and the emission per delivery is reduced. Also, it is important to see the trend, because then it is bogged down more, and in some cases increase again slightly, because the distance and the capacity mark the trend. 88 David Bardasco San José Figure 10: Emissions Case 1 The case 25 follows the same trend as above, and despite to have more members of the Crowd, such as has been said previously, in the model raised this does not affect. Neither a greater number of stores. Therefore, as in the previous case, what marking the trend is guided by the distance travelled and the capacity of the transport. Figure 11: Emissions Case 25 In conclusion, after analysing the proposed model, it can be said that the results obtained are consistent and are mainly governed by the distance travelled and by the transport used, which depending on the emissions that emits is what Crowdsourcing Logistics in B2C e-Commerce 89 determines the pollution deliveries, and in turn of the practice cited. It is important to say the opportunity, possibility to get make a completely clean transport, but on the contrary it implies increased fatigue by the member of the Crowd and more time. 96 David Bardasco San José www.guardian.co.uk/environment/2007/sep/12/ plasticbags.supermarkets (accessed 6 December 2007) So, H. W., Gunasekaran, A., & Chung, W. W. (2006). Last Mile fulfilment strategy for competitive advantage. International Journal of Logistics Systems and Management, 2(4), 404-418. Srivastava, Samir K. 2007. “Green Supply-Chain Management: A State-of-theArt Literature Review.” International Journal of Management Reviews 9(1): 53–80. Vanelslander, T., Deketele, L., & Van Hove, D. (2013). Commonly used ecommerce supply chains for fast moving consumer goods: comparison and suggestions for improvement. International Journal of Logistics Research and Applications, 16(3), 243-256. Visser, J.G.S.N & T. Nemoto (2001), E-commerce and the Consequences for Freight Transport. Research report, Ministry of Economic Affairs. Wei, Z. and Zhou, L. (2011), “Case study of online retailing fast fashion industry”, International Journal of e-Education, e-Business, e-Management and e-Learning, Vol. 1 No. 3, pp. 195-200. Weltevreden, J. W. (2008). B2c e-commerce logistics: the rise of collection-anddelivery points in The Netherlands. International Journal of Retail & Distribution Management, 36(8), 638-660. Xu, M., Ferrand, B., & Roberts, M. (2008). The last mile of e-commerceunattended delivery from the consumers and eTailers' perspectives. International Journal of Electronic Marketing and Retailing, 2(1), 20-38. Crowdsourcing Logistics in B2C e-Commerce 97 6.2. WEBS · AMAZON FLEX: - https://flex.amazon.com/ - http://www.manutencionyalmacenaje.com/es/notices/2016/01/amazo n-flex-ultima-milla-reparto-de-paquetes-al-estilo-uber38753.php#.VvHpZT-qzL8 - http://www.xternaliza.es/amazon-tiene-previsto-la-expansion-delproyecto-flex/ - http://www.tuexperto.com/2015/09/30/amazon-pagara-a-los-usuariospor-repartir-paquetes/ - http://www.logisticaytransporte.es/noticias.php/Amazon-afianza-sunuevo-servicio-de-entregas-en-colaboraci%C3%B3n-conconductores-aut%C3%B3nomos.-cl.--amazonpaqueter%C3%ADa/61399 - http://www.businessinsider.com/amazon-working-on-secretiveproject-called-amazon-flex-2015-8 - http://www.geekwire.com/2015/amazon-set-to-launch-new-amazonflex-package-pickup-service-in-seattle-area-with-prime-now/ - https://en.wikipedia.org/wiki/Amazon.com · DELIV: - https://www.deliv.co/ - https://www.crunchbase.com/organization/deliv#/entity - http://www.cnet.com/news/speedy-delivery-startup-deliv-expands-tonine-new-markets/ - https://en.wikipedia.org/wiki/Deliv - http://www.allthefrugalladies.com/shopping/become-deliv-driver/ · DOORDASH: - https://www.doordash.com/ · INSTACART: 98 David Bardasco San José - https://www.instacart.com - http://www.thepennyhoarder.com/dont-mind-grocery-shopping-make25-per-hour-delivering-food-with-instacart/ · MyWays: - https://en.wikipedia.org/wiki/DHL_Express - https://www.myways.com/ - http://www.dhl.com/en/press/releases/releases_2013/logistics/dhl_c rowd_sources_deliveries_in_stockholm_with_myways.html#.VwZw Jj_rzL8 - http://www.delivered.dhl.com/en/articles/2014/04/what-s-the-storymr-oom.html · POSTAMATES: - https://postmates.com/ - https://en.wikipedia.org/wiki/Postmates · RICKSHAW: - https://gorickshaw.com/ - https://angel.co/rickshaw/activity · SIDECAR DELIVERIES: - https://www.side.cr/ - https://en.wikipedia.org/wiki/Sidecar_%28company%29 - https://www.theguardian.com/technology/2015/dec/30/ubercompetitor-sidecar-shut-down - https://www.side.cr/why-we-sold-to-gm/ · UberEATS: - https://ubereats.com - http://postandparcel.info/71869/news/ubereats-sets-to-expand/ · UberRUSH: - http://www.logisticaytransporte.es/noticias.php/Uber-lanza-su-propiaempresa-de-paqueter%C3%ADa.-cl.--paqueteria/62627 - https://rush.uber.com/how-it-works Crowdsourcing Logistics in B2C e-Commerce 99 - https://www.uber.com - https://en.wikipedia.org/wiki/Uber_%28company%29 - https://newsroom.uber.com/us-new-york/a-reliable-ride-for-yourdeliveries/ - http://lyftubernewsletter.com/uber-rush/ - http://www.jobmonkey.com/shared-economy/on-demanddelivery/uber-rush/ · ZIMPMENTS: - https://zipments.com/ - http://www.xconomy.com/new-york/2015/11/10/deliv-acquireszipments-for-undisclosed-sum/ - https://angel.co/zipments/activity http://www.crowdsourcing.org/editorial/delivery-gets-personal-withzipments/27808 100 David Bardasco San José Crowdsourcing Logistics in B2C e-Commerce 101 ANNEXES 1. Results of the simulation RESULTS SCENARIO AND CASES NAME OF CASE TRANSPORT COSTS EMISSIONS CASE 1-A By CAR Petrol 7,93 1,13 By CAR Diesel 7,36 1,04 By MOTORBIKE 4,69 0,89 By BICYCLE 5,40 0,00 On FOOT 14,16 0,00 CASE 1-B By CAR Petrol 4,46 0,58 By CAR Diesel 4,17 0,53 By MOTORBIKE 2,81 0,46 By BICYCLE 3,17 0,00 On FOOT 7,64 0,00 CASE 1-C By CAR Petrol 3,80 0,48 By CAR Diesel 3,56 0,44 By MOTORBIKE 2,43 0,38 By BICYCLE 3,32 0,00 On FOOT 8,15 0,00 CASE 1-D By CAR Petrol 3,37 0,41 By CAR Diesel 3,17 0,38 By MOTORBIKE 2,69 0,44 By BICYCLE 3,37 0,00 On FOOT 8,29 0,00 CASE 1-E By CAR Petrol 3,37 0,41 By CAR Diesel 3,16 0,38 By MOTORBIKE 2,69 0,44 By BICYCLE 3,74 0,00 On FOOT 9,38 0,00 CASE 2-A By CAR Petrol 13,98 1,13 By CAR Diesel 13,41 1,04 By MOTORBIKE 10,38 0,89 By BICYCLE 11,45 0,00 On FOOT 20,21 0,00 CASE 2-B By CAR Petrol 14,12 1,15 By CAR Diesel 13,54 1,06 By MOTORBIKE 10,81 0,92 By BICYCLE 11,54 0,00 On FOOT 20,47 0,00 CASE 2-C By CAR Petrol 9,89 0,48 By CAR Diesel 9,65 0,44 By MOTORBIKE 8,52 0,38 102 David Bardasco San José By BICYCLE 9,41 0,00 On FOOT 14,24 0,00 CASE 2-D By CAR Petrol 9,46 0,41 By CAR Diesel 9,26 0,38 By MOTORBIKE 8,78 0,44 By BICYCLE 9,46 0,00 On FOOT 14,38 0,00 CASE 2-E By CAR Petrol 9,46 0,41 By CAR Diesel 9,25 0,38 By MOTORBIKE 9,05 0,50 By BICYCLE 9,83 0,00 On FOOT 15,47 0,00 CASE 3-A By CAR Petrol 20,03 1,13 By CAR Diesel 19,46 1,04 By MOTORBIKE 16,79 0,90 By BICYCLE 17,50 0,00 On FOOT 26,26 0,00 CASE 3-B By CAR Petrol 20,17 1,15 By CAR Diesel 19,59 1,06 By MOTORBIKE 16,86 0,92 By BICYCLE 17,59 0,00 On FOOT 26,52 0,00 CASE 3-C By CAR Petrol 15,94 0,48 By CAR Diesel 15,70 0,44 By MOTORBIKE 14,57 0,38 By BICYCLE 15,46 0,00 On FOOT 20,29 0,00 CASE 3-D By CAR Petrol 15,51 0,41 By CAR Diesel 15,31 0,38 By MOTORBIKE 14,83 0,44 By BICYCLE 15,51 0,00 On FOOT 20,43 0,00 CASE 3-E By CAR Petrol 15,51 0,41 By CAR Diesel 15,30 0,38 By MOTORBIKE 15,10 0,50 By BICYCLE 15,88 0,00 On FOOT 21,52 0,00 CASE 4-A By CAR Petrol 26,08 1,13 By CAR Diesel 25,51 1,40 By MOTORBIKE 22,84 0,90 By BICYCLE 23,55 0,00 On FOOT 32,31 0,00 CASE 4-B By CAR Petrol 26,22 1,15 By CAR Diesel 25,64 1,06 By MOTORBIKE 22,91 0,92 Crowdsourcing Logistics in B2C e-Commerce 103 By BICYCLE 23,64 0,00 On FOOT 32,57 0,00 CASE 4-C By CAR Petrol 21,99 0,48 By CAR Diesel 21,75 0,44 By MOTORBIKE 20,62 0,38 By BICYCLE 21,51 0,00 On FOOT 26,34 0,00 CASE 4-D By CAR Petrol 21,56 0,41 By CAR Diesel 21,36 0,38 By MOTORBIKE 20,88 0,44 By BICYCLE 21,56 0,00 On FOOT 26,48 0,00 CASE 4-E By CAR Petrol 21,56 0,41 By CAR Diesel 21,35 0,38 By MOTORBIKE 21,15 0,50 By BICYCLE 21,55 0,00 On FOOT 26,47 0,00 CASE 5-A By CAR Petrol 32,13 1,13 By CAR Diesel 31,56 1,04 By MOTORBIKE 28,89 0,90 By BICYCLE 29,60 0,00 On FOOT 38,36 0,00 CASE 5-B By CAR Petrol 32,27 1,15 By CAR Diesel 31,69 1,06 By MOTORBIKE 28,96 0,92 By BICYCLE 29,69 0,00 On FOOT 38,62 0,00 CASE 5-C By CAR Petrol 28,04 0,48 By CAR Diesel 27,80 0,44 By MOTORBIKE 26,67 0,38 By BICYCLE 27,56 0,00 On FOOT 32,39 0,00 CASE 5-D By CAR Petrol 27,61 0,41 By CAR Diesel 27,41 0,38 By MOTORBIKE 26,93 0,44 By BICYCLE 27,61 0,00 On FOOT 32,53 0,00 CASE 5-E By CAR Petrol 27,61 0,41 By CAR Diesel 27,40 0,38 By MOTORBIKE 27,20 0,50 By BICYCLE 27,60 0,00 On FOOT 32,52 0,00 CASE 6-A By CAR Petrol 7,90 1,13 By CAR Diesel 7,34 1,04 By MOTORBIKE 4,66 0,90 104 David Bardasco San José By BICYCLE 5,37 0,00 On FOOT 14,14 0,00 CASE 6-B By CAR Petrol 4,43 0,58 By CAR Diesel 4,15 0,53 By MOTORBIKE 2,78 0,46 By BICYCLE 3,14 0,00 On FOOT 7,61 0,00 CASE 6-C By CAR Petrol 3,80 0,48 By CAR Diesel 3,56 0,44 By MOTORBIKE 2,43 0,38 By BICYCLE 3,32 0,00 On FOOT 8,15 0,00 CASE 6-D By CAR Petrol 3,37 0,41 By CAR Diesel 3,17 0,38 By MOTORBIKE 2,69 0,44 By BICYCLE 3,37 0,00 On FOOT 8,29 0,00 CASE 6-E By CAR Petrol 3,37 0,41 By CAR Diesel 3,16 0,38 By MOTORBIKE 2,69 0,44 By BICYCLE 3,74 0,00 On FOOT 9,38 0,00 CASE 7-A By CAR Petrol 7,93 1,13 By CAR Diesel 7,36 1,04 By MOTORBIKE 4,69 0,89 By BICYCLE 5,40 0,00 On FOOT 14,16 0,00 CASE 7-B By CAR Petrol 4,43 0,58 By CAR Diesel 4,15 0,53 By MOTORBIKE 2,78 0,46 By BICYCLE 3,14 0,00 On FOOT 7,61 0,00 CASE 7-C By CAR Petrol 3,80 0,48 By CAR Diesel 3,56 0,44 By MOTORBIKE 2,43 0,38 By BICYCLE 3,32 0,00 On FOOT 8,15 0,00 CASE 7-D By CAR Petrol 3,37 0,41 By CAR Diesel 3,17 0,38 By MOTORBIKE 2,69 0,44 By BICYCLE 3,37 0,00 On FOOT 8,29 0,00 CASE 7-E By CAR Petrol 3,37 0,41 By CAR Diesel 3,16 0,38 By MOTORBIKE 2,69 0,44 Crowdsourcing Logistics in B2C e-Commerce 105 By BICYCLE 3,74 0,00 On FOOT 9,38 0,00 CASE 8-A By CAR Petrol 13,95 1,13 By CAR Diesel 13,39 1,04 By MOTORBIKE 10,71 0,90 By BICYCLE 11,42 0,00 On FOOT 20,19 0,00 CASE 8-B By CAR Petrol 14,09 1,15 By CAR Diesel 13,52 1,06 By MOTORBIKE 10,79 0,92 By BICYCLE 11,51 0,00 On FOOT 20,45 0,00 CASE 8-C By CAR Petrol 9,87 0,48 By CAR Diesel 9,63 0,44 By MOTORBIKE 8,50 0,38 By BICYCLE 9,38 0,00 On FOOT 14,21 0,00 CASE 8-D By CAR Petrol 9,44 0,41 By CAR Diesel 9,23 0,38 By MOTORBIKE 8,76 0,44 By BICYCLE 9,43 0,00 On FOOT 14,35 0,00 CASE 8-E By CAR Petrol 21,56 0,41 By CAR Diesel 21,35 0,38 By MOTORBIKE 21,15 0,50 By BICYCLE 21,93 0,00 On FOOT 27,57 0,00 CASE 9-A By CAR Petrol 19,98 1,13 By CAR Diesel 19,41 1,04 By MOTORBIKE 16,74 0,90 By BICYCLE 17,45 0,00 On FOOT 26,21 0,00 CASE 9-B By CAR Petrol 20,12 1,15 By CAR Diesel 19,54 1,06 By MOTORBIKE 16,81 0,92 By BICYCLE 17,54 0,00 On FOOT 26,47 0,00 CASE 9-C By CAR Petrol 15,89 0,48 By CAR Diesel 15,65 0,44 By MOTORBIKE 14,52 0,38 By BICYCLE 15,41 0,00 On FOOT 20,24 0,00 CASE 9-D By CAR Petrol 15,46 0,41 By CAR Diesel 15,26 0,38 112 David Bardasco San José By MOTORBIKE 8,49 0,38 By BICYCLE 9,37 0,00 On FOOT 14,20 0,00 CASE 20-D By CAR Petrol 9,42 0,41 By CAR Diesel 9,22 0,34 By MOTORBIKE 8,75 0,44 By BICYCLE 9,42 0,00 On FOOT 14,34 0,00 CASE 20-E By CAR Petrol 9,42 0,41 By CAR Diesel 9,21 0,38 By MOTORBIKE 9,02 0,50 By BICYCLE 9,79 0,00 On FOOT 15,43 0,00 CASE 21-A By CAR Petrol 7,89 1,13 By CAR Diesel 7,32 1,04 By MOTORBIKE 4,65 0,90 By BICYCLE 5,36 0,00 On FOOT 14,12 0,00 CASE 21-B By CAR Petrol 4,42 0,58 By CAR Diesel 4,13 0,53 By MOTORBIKE 2,77 0,46 By BICYCLE 3,13 0,00 On FOOT 7,60 0,00 CASE 21-C By CAR Petrol 3,80 0,48 By CAR Diesel 3,56 0,44 By MOTORBIKE 2,43 0,38 By BICYCLE 3,32 0,00 On FOOT 8,15 0,00 CASE 21-D By CAR Petrol 3,37 0,41 By CAR Diesel 3,17 0,38 By MOTORBIKE 2,69 0,44 By BICYCLE 3,37 0,00 On FOOT 8,29 0,00 CASE 21-E By CAR Petrol 3,37 0,41 By CAR Diesel 3,16 0,38 By MOTORBIKE 2,69 0,44 By BICYCLE 3,74 0,00 On FOOT 9,38 0,00 CASE 22-A By CAR Petrol 7,89 1,13 By CAR Diesel 7,32 1,04 By MOTORBIKE 4,65 0,90 By BICYCLE 5,36 0,00 On FOOT 14,12 0,00 CASE 22-B By CAR Petrol 4,42 0,58 By CAR Diesel 4,13 0,53 Crowdsourcing Logistics in B2C e-Commerce 113 By MOTORBIKE 2,77 0,46 By BICYCLE 3,13 0,00 On FOOT 7,60 0,00 CASE 22-C By CAR Petrol 3,80 0,48 By CAR Diesel 3,56 0,44 By MOTORBIKE 2,43 0,38 By BICYCLE 3,32 0,00 On FOOT 8,15 0,00 CASE 22-D By CAR Petrol 3,37 0,41 By CAR Diesel 3,17 0,38 By MOTORBIKE 2,69 0,44 By BICYCLE 3,37 0,00 On FOOT 8,29 0,00 CASE 22-E By CAR Petrol 3,37 0,41 By CAR Diesel 3,16 0,38 By MOTORBIKE 2,69 0,44 By BICYCLE 3,74 0,00 On FOOT 9,38 0,00 CASE 23-A By CAR Petrol 7,93 1,13 By CAR Diesel 7,36 1,04 By MOTORBIKE 4,69 0,90 By BICYCLE 5,40 0,00 On FOOT 14,16 0,00 CASE 23-B By CAR Petrol 8,07 1,15 By CAR Diesel 7,49 1,06 By MOTORBIKE 4,76 0,92 By BICYCLE 5,49 0,00 On FOOT 14,24 0,00 CASE 23-C By CAR Petrol 3,80 0,48 By CAR Diesel 3,56 0,44 By MOTORBIKE 2,43 0,38 By BICYCLE 3,32 0,00 On FOOT 8,15 0,00 CASE 23-D By CAR Petrol 3,37 0,41 By CAR Diesel 3,17 0,38 By MOTORBIKE 2,69 0,44 By BICYCLE 3,37 0,00 On FOOT 8,29 0,00 CASE 23-E By CAR Petrol 3,37 0,41 By CAR Diesel 3,16 0,38 By MOTORBIKE 2,69 0,44 By BICYCLE 3,74 0,00 On FOOT 9,38 0,00 CASE 24-A By CAR Petrol 7,93 1,13 By CAR Diesel 7,36 1,04 114 David Bardasco San José By MOTORBIKE 4,69 0,90 By BICYCLE 5,40 0,00 On FOOT 14,16 0,00 CASE 24-B By CAR Petrol 8,07 1,15 By CAR Diesel 7,49 1,06 By MOTORBIKE 4,76 0,92 By BICYCLE 5,49 0,00 On FOOT 14,24 0,00 CASE 24-C By CAR Petrol 3,80 0,48 By CAR Diesel 3,56 0,44 By MOTORBIKE 2,43 0,38 By BICYCLE 3,32 0,00 On FOOT 8,15 0,00 CASE 24-D By CAR Petrol 3,37 0,41 By CAR Diesel 3,17 0,38 By MOTORBIKE 2,69 0,44 By BICYCLE 3,37 0,00 On FOOT 8,29 0,00 CASE 24-E By CAR Petrol 3,37 0,41 By CAR Diesel 3,16 0,38 By MOTORBIKE 2,69 0,44 By BICYCLE 3,74 0,00 On FOOT 9,38 0,00 CASE 25-A By CAR Petrol 7,93 1,13 By CAR Diesel 7,36 1,04 By MOTORBIKE 4,69 0,90 By BICYCLE 5,40 0,00 On FOOT 14,16 0,00 CASE 25-B By CAR Petrol 8,07 1,15 By CAR Diesel 7,49 1,06 By MOTORBIKE 4,76 0,92 By BICYCLE 5,49 0,00 On FOOT 14,24 0,00 CASE 25-C By CAR Petrol 3,80 0,48 By CAR Diesel 3,56 0,44 By MOTORBIKE 2,43 0,38 By BICYCLE 3,32 0,00 On FOOT 8,15 0,00 CASE 25-D By CAR Petrol 3,37 0,41 By CAR Diesel 3,17 0,38 By MOTORBIKE 2,69 0,44 By BICYCLE 3,37 0,00 On FOOT 8,29 0,00 CASE 25-E By CAR Petrol 3,37 0,41 By CAR Diesel 3,16 0,38 Crowdsourcing Logistics in B2C e-Commerce 115 By MOTORBIKE 2,69 0,44 By BICYCLE 3,74 0,00 On FOOT 9,38 0,00 · Note: The total distance considered between the shop and the customers are: - 1 customer: 9,72 km. - 2 customers: 9,94 km. - 3 customers: o On foot and by bike: 16,97 km. o By car and by motorbike: 12,70 km. - 4 customers: o On foot and by bike: 17,33 km. o By Motorbike: 14,96 km. o By Car: 10,68 km. - 4 customers: o On foot and by bike: 20,05 km. o By Motorbike: 17,307 km. o By Car: 10,65 km. The distance considered between the member of the Crowd and the Shop 2 is: 1,37 km. 116 David Bardasco San José