Economic Evaluation of RFID Technology in the Production Environment

1.1. RFID advantages

Generally speaking, the current RFID advantages with respect to the commonly used barcode, and also well covered in literature, are:

Asset/product tracking. RFID can be used to monitor and manage the physical movement of materials and finished products, to generate and deliver a flow of critical information, and to form the basis for enhanced working between supply chain partners in real time. This capability enhances the control of operations by reducing stock losses and improving supply chain visibility and integration. Nowadays supply chain integration is considered strategic in order to obtain a competitive advantage [1]. Non-conventional objects for tracking were applied in the Taipei Medical University Hospital: here this technology has been used for a location-based medical service system. In this framework its main function was to monitor physiological signals and track the paths of infection sources.

However, this technology and its comparison with barcode scanning, even if confirming the net technical advantages of the former, still needs to be exhaustively researched due to the wide field of application and potentialities [17].

1.2. Bibliographic analysis of evaluation models

The bibliographic analysis focused its attention on the most recent and significant models dealing with the RFID technology investment evaluation. Backer et al. [18] developed a multi-perspective model for the measurement of the investment value in the RFID technology applicable to the entire supply chain (SC). The SC actors: suppliers, logistic operators, manufacturers, retailers, customers, etc., represent the highest hierarchical level of the model. Starting from this level, it is possible to identify the principal processes characterizing every actor and subsequently decompose them into more and more detailed activities, until coming to the processes directly involved in the RFID technology introduction. Each actor, from the logistics side, can be involved in seven main processes: procurement, production, redemption, transport, customer service, delivery and return. Main processes are only the highest level, they can be decomposed into more detailed sub-levels up to identifying activities directly involved in the RFID technology application. To each lowest level activity are associated the so-called “process objects”: product (level 0), packaging (level 1), transport unit (level 2), shipment unit (level 3), and container or vehicle (level 5). Moreover, for each lowest level activity, one or more effects can be identified caused by the eventual introduction of RFID technology: process time reduction, error reduction, resource consumption reduction, and process information (traceability). Once we have identified the processes affected by the technology and theirs technical and economical impact through the four effects previously illustrated, it is necessary to compare the actual state (AS-IS) with the future one (TO-BE) in terms of costs.

Ustundag et al. [19], instead, proposed a model for the evaluation of RFID technology investment using a FRBS (Fuzzy Rule-based System) for the assessment of the demand increase due to the introduction of this new technology. The model identifies the elements that contribute to the definition of the investment cost and the differential cost and revenue that can be, positively or negatively, influenced by the new technology. The use of fuzzy logic allows us to estimate the increase of demand depending on the implementation choices through rules defined by sector experts. The Montecarlo simulation ultimately allows us to vary the parameters inside the evaluation ranges and to obtain a more reliable NPV value.

De Marco et al. [20] proposed a System Dynamics (SD) model for the assessment of the RFID technology impact on the operations of a retail store of a well-known fashion store chain. The SD model gives the evaluation of the values concerning time reduction for product receiving, inventorying, labelling, etc., and concerning sales increasing due to the increase of time that the personnel can dedicate to client service. These records represent the inputs for the CFA (Cash Flow Analysis). This work relates to ex-ante evaluation of RFID technology impact on supply chain performances and, in particular, on the last ring of this chain, the retail store, which is often less well analysed in the literature.

Bottani et al. [21] produced a work on the qualitative evaluation of the economic RFID impact on a three-level chain composed of a producer DC, a distributor and a retailer. The study was undertaken by presenting a questionnaire to a group of leading companies in the Fast Moving Consumer Good (FMCG) sector. The results showed that use of a RFID system was convenient for all the chain actors when tagging pallets. When tagging cases, instead, the advantages for distributor and retailer do not compensate for the higher costs incurred by the manufacturer.

Bottani [22] proposed an economic evaluation model for the assessment of the impact in the adoption of RFID technology on the traceability management in the supply chain. The tags, applied at case level, allow reducing the recall costs of a defective lot through the distribution chain. This model does not consider the implementation cost of the RFID technology and it is supposed that the technology is already implemented. The effects generated concern recall point number reduction and overall recall goods reduction. Those effects produce reductions in transport costs, increased sales (due to a reduction in recalled goods), fewer product disposal costs and reductions in the costs associated with the traceability of goods. The work presents an interesting mathematical formulation for the definition of each of the above-mentioned costs.

Bertolini et al. [23], instead, analysed the RFID technology impact on the fashion industry's supply chain. Benefits are evaluated both in economic terms and in terms of the added value produced by greater information availability due to the application of tags at item level. The study was conducted at 13 clothing industry companies and in three different phases. The new proposed scenario was tested through a test project and showed manpower productivity increases (goods shipment and receiving), greater accuracy in inventory, higher speed in inventory operations, a precise traceability of clothes tried on by customers and, ultimately, a turnover increase and sales and unsold stock real time monitoring.

Dai and Tseng [24] studied the RFID impact on greater stock accuracy in a multi-stadium supply chain underlining a substantial associated cost reduction. The model considers only shrinkage errors, but the authors indicated that the eventual introduction of the model affected factors such as misplacement, transaction errors, etc., and could further increase the impact on stocks and costs.

Busato et al. [25] analysed the opportunity of introducing RFID in a large European Consumer Packaged Goods (FMCG) company in the cosmetics and beauty sector. Eleven processes in receiving, handling, storage and delivery procedures in one of the main distribution centres of the company were analysed and modelled using a detailed business process modelling approach. Since savings were evaluated only in terms of the reduction of resources used - achieved through process acceleration - despite the company dealing in huge volumes of high-margin products, the feasibility study turned out to be negative.

The preceding works were selected because they offer a clear picture of the RFID technology investment evaluation and because they integrate themselves, thereby creating an almost complete framework able to face the problem at each level and with all instruments: problem classification and fragmentation, identification of cost voices involved, dynamic model for the assessment of casual factors influenced by the technology. In the next paragraph, using the described models, developing and conveniently integrating them, we will present the complete framework for the assessment of the RFID technology investment for a manufacturer.

2.1. Standard framework

The standard framework allows the analyst, through schematizing processes, activities and standard objects valid for every productive contest, to define the macro-activities to be exploded in order to identify those influenced by RFID technology. To achieve this objective, the implementation project's goals have to be clearly defined.

2.1.3. Macro-activities

The framework, compared with Becker's model (2010), describes manufacturer's specific main processes' phases in more detail. In particular, for each main process, macro-activities more influenced by RFID technology are reported in table 1. They are indicated as A_ij (j-th macro-activity of i-th process).

OPEN IN VIEWER

Table 1.

Macro-activities

Ai Main Processes	Aik Macro-activities
A1 Procurement	A11 Raw Materials Entrance
	A12 Quality Control
	A13 Labelling
	A14 Materials Storage
	A16 Returns Mgnt
A2 Production	A21 Materials Entrance
	A22 Process Execution
	A23 Semi-finished Products Localization
	A24 Products Exit
	A25 Facilities Localization
	A26 Maintenance
A3 Redemption	A31 Returns Mgnt
A4 Transport	A41 Raw Materials Mov.
	A42 Facilities Mov.
	A43 Semi-finished Products Mov.
	A44 Finished Products Mov.
	A45 External Suppliers Mov.
A5 Customer Service	A51 Lots Identification
	A52 Products and Components Identification
	A53 Spare Parts Management
A6 Delivery	A61 Transport Unit Preparation
	A62 Picking
	A63 Destination Control
	A64 Transport Facilities' Mgnt
A7 Return	A71 Transport Facilities' Mgnt
A8 Inventory	A81 Entrance
	A82 Exit
	A83 Conservation
	A84 Material Localization

Activities reported in the previous chart, depending on the specific company, will be further detailed during the application, with the purpose of identifying micro phases that are influenced by RFID systems' use.

2.1.4. Process object

The activities' identification, obviously, depends on the RFID tag application level that the company decides to implement. Clearly, activities are different if the tag is applied at the item level (product) or at the container level. Process objects are classified by Becker and levels are indicated with h index (see table 2).

OPEN IN VIEWER

Table 2.

Process Object

Level (h)	Process Object
0	Item
1	Packaging
2	Transport Unit
3	Delivery Unit
4	Container
5	Vehicle

2.2. Application assessment

The application's assessment phase is different from the previous one, standard framework, because it can be realized only by considering the peculiarity of processes and activities influenced by re-engineering through RFID technology. In fact, micro-activities that involve processes and manufacturing phases directly influenced by RFID technology can be defined only by knowing specific characteristics of the company involved. The application's assessment is divided into two phases: technical analysis, that compares, from a technical point of view, the existing scenario with the re-engineered one; and economic assessment, that translates the technical analysis in an economic/financial comparison.

2.3. Economic assessment

2.3.1. Operational costs

This phase allows us to translate the technological impacts previously defined into cost terms. Knowing activities and produced effects, it is possible to calculate costs and/or earnings inducted variations. Cost voices considered are the ones proposed by Ustundag [19] to which is added the voice related to maintenance cost variation (ΔC_M). The mathematical formulation to compute the single value variation is reported as follows. They differ from the ones proposed by Ustundag due to the different nature of the considered elements. In fact, it is not present earnings increase, because this effect is considered irrelevant for manufacturers, but the voice related to maintenance variation costs that is introduced. Variation costs are:

1. Work Cost Variation (ΔC_L)

This cost voice is a function of PTR and ER, because both effects can induce a reduction in the time needed to perform operations and, so, induce a reduction in workforce, with the same number of finished products manufactured.

Δ C L = ∑ i , j C L i j F T E ⋅ ∑ k , h μ ( A i j ( h ) ) ⋅ ( P T R ( A i j k ( h ) ) + + E R ( A i j ( h ) ) )

(1)

Where FTE is Full Time Equivalent (a person who works 8 hours a day for 218 days a year, 1 744 hours/year), CL_ij is the average work cost of j-th macro-activity of i-th process, μ ( A i j ( h ) ) is the average value of A i j ( h ) macro-activity throughput and T ( A i j ( h ) ) is the average time needed to perform the activity A i j ( h ) for which error reduction is performed.

2. Inventory Cost Variation (ΔC_I)

Stocks can be influenced by operation performing time reduction. As seen before, this reduction is principally due to PTR and ER effects. Moreover, inventory entity is also influenced by a more rapid process information PI. In fact, while in the first case the reduction is well known thank to Little's law, in the second the eventual process information lateness reduction reduces virtual safety stock related to it. In other words, purchasing a supply order a day later is equivalent to a safety stock increase, with the same service level offered to the client.

Δ C I = ∑ i , j C L i j ⋅ ( ∑ k , h μ ( A i j ( h ) ) ⋅ ( P T R ( A i j k ( h ) ) + + E R ( A i j ( h ) ) ) + ∑ h σ ( A i j ( h ) ) ⋅ P I ( A i j ( h ) ) )

(2)

Where CI_ij is the j-th macro-activity average cost of storage, μ and σ are respectively average value and standard deviation of A i j ( h ) macro-activity throughput.

3. Shortage Cost Variation (ΔC_S)

Shortage cost variation is an additional effect of process information and process execution time reduction.

Δ C S = ∑ i , j C S i j ⋅ ∑ h ( μ ( A i j ( h ) ) ⋅ E R ( A i j ( h ) ) + + σ ( A i j ( h ) ) ⋅ P I ( A i j ( h ) ) )

(3)

Where CS_ij is the unitary cost of j-th macro-activity components shortage. ΔC_I e ΔC_S computation needs a clear division between inventory and stock-out effects, in order to avoid an overlapping of the two events and a consequent cost increase.

4. Maintenance Cost Variation (ΔC_M)

A positive variation of such a cost is due to RFID technology presence (antennas, readers, tags, etc.), vice versa, the production and movement facilities' decrease reduces this cost voice.

Δ C M = ∑ i , j ( ∑ h C M i j ⋅ μ ( A i j ( h ) ) ⋅ ∑ k P T R ( A i j k ( h ) ) ⋅ γ ( A i j k ( h ) ) + R C R ( A i j ( h ) ) ⋅ δ ( A i j ( h ) ) ) - + ∑ i , j , h μ ( A i j ( h ) ) ⋅ α ( A i j ( h ) ) ⋅ C T G ( h ) - C H D ⋅ β

(4)

Where CM_ij is the average maintenance cost for j-th macro-activity resources, γ ( A i j k ( A ) ) is the percentage of PTR that involves physical resources, γ ( A i j k ( A ) ) is the percentage of maintenance cost, CTG^(h) is the tag cost and δ ( A i j ( h ) ) is the tag annual substitution percentage, CHD is RFID system hardware cost and β is the average percentage value of maintenance cost incidence.

The overall annual cost variation is given by previous values' sum.

Δ C T = Δ C L + Δ C I + Δ C S + Δ C M

(5)

2.3.2. Implementation costs

Implementation costs, instead, are defined considering macro-classes defined by Ustundag et al. [19].

C_x Hardware Costs. RFID system implementation requires a series of hardware components that can be gathered into the following classes:

Middleware

C_y Software costs. These costs include the software license purchase necessary for RFID system operation and interface implementation for the connection to existing management systems.

C_z Other Costs. Integration, training, update, etc.

Hardware related costs can also be partly identified for each macro-activity because facilities depend on the process in which they insert themselves. Common costs of facilities are software developing/purchase costs and existing management systems integration and update costs - training costs, instead, can be defined as a whole (eq. 6).

C H D = C x + C y + C z

(6)

2.3.3. Profitability analysis

To perform economic convenience analysis, naming as N the number of years for the analysis, as t the index related to the year for which the cash flow is considered and as i the discounting rate, the used evaluation indicators are:

Net Present Value (NPV)

N P V = C H D + ∑ t = 0 N Δ C T ( t ) ( 1 + i ) t

(7)

Internal Rate of Return (IRR)

I R R → C H D + ∑ t = 0 N Δ C T ( t ) ( 1 + I R R ) t = 0

(8)

Pay Back Period (PBP)

P B P → C H D + ∑ t = 0 P B P Δ C T ( t ) = 0

(9)

3.1. Framework of the case study

A CPG (Consumer Packaged Goods) industry or FMCG (Fast Moving Consumer Goods) industry is one which works with products that are sold quickly and at relatively low cost. Examples of this industry type are food and beverages, footwear and clothing, tobacco and cleaning products. Indeed, the case study proposed in this paper relates to an industry within the food sector.

3.1.1. Problem statement

The need for RFID implementation came from an internal project which identified this technology as necessary in order to improve the already present goods Tracking & Tracing system (T&T), whose scope was to comply the updated normative within the food sector. Indeed, the more stringent normative has pushed the company to choose a more evolved system in order to reduce costs, comply with the higher level requirements and gain in company reputation. For the aforementioned reasons, RFID thus became the mainstay of the whole selected system. The implementation of the new T&T system, after a high level analysis, has been tested in the real company through a pilot project in a limited sub-process. In the following paragraphs, the results of this study have been reported utilizing the evaluation model discussed before.

3.1.1.1 Main processes, macro-activities and process object

The factory zone in which the project implementation is imagined is represented in figure 2.

OPEN IN VIEWER

Figure 2.

Factory Layout

From the layout it is possible to identify six main areas:

Production Hall 1 (PH1) in which pasta, vegetables and sauce semi-finished products are manufactured;

Production Hall 2 (PH2) that performs packaging for finished products and for some semi-finished products that will be later treated to a complete manufacturing process;

Production Hall 3 (PH3) that realizes finished products;

Raw Material and packaging warehouse (RM);

Coldstore (CS1) of raw materials and semi-finished products;

Coldstore (CS2) of finished products.

Referring to the previous definitions, the implementation of the T&T system and the improvement of internal logistic efficiency with a consequent reduction of material research time, and related costs, and the definition of new standards concerning semi-finished products turn-around (movement) from coldstores to production halls (PH) allows us to confine the application to two main processes:

Production (A₂);

Inventory (A₈);

Macro-activities involved and connected process objects (pallet transport unit) are indicated in table 3.

OPEN IN VIEWER

Table 3.

Macro-activities of the case study

Ai Main Processes	A(h) ik Macro-activities
A2 Production	A(2)21 Materials Entrance
	A(2)24 Products Exit
A8 Inventory	A(2)81 Entrance
	A(2)82 Exit

3.2. Application assessment

A business process analysis was carried in order to identify the processes mainly involved in reaching the objectives.

3.2.1. Micro-activities and impact analysis

AS-IS

The following AS-IS analysis diagram indicates factory processes and activities currently in use.

All transfer orders confirmation phases and pallet creation phases are performed manually by one or more dedicated operators.

It is not available information about pallets movement between production and warehouses.

Aggregate or punctual statistics concerning pallet movements with regard to traceability and products quality control are not available automatically.

TO-BE

Coherent with company targets, desired states were exploded. They are:

Complete automation of product identification and tracking of the route between production departments and warehouses and vice versa.

Reduction of registration time, pallet control and transfer from coldstores to production points.

Reduction in eventual pallet identification errors that are frequent in the actual procedures.

Based on above-mentioned points, the proposed solution, using RFID technology, will have to trace pallet movements and to automate production processes, modifying them in order to:

Automatically identify pallets and acquire weight in the pallet creation station, automatically create pallet in SAP software contextually opening a transfer order to the coldstore, after having printed fiche to put on the pallet

Automatically identify the pallets crossing the coldstore entrance and activate transfer order confirmation in SAP

Automatically identify the pallets coming out from the coldstore and activate in SAP a transfer order confirmation to the production department or other coldstore

Automatically identify the pallets on hoppers for production entrance

Micro-activities influenced by RFID technology application (see Table 4 – the process object (2) is not shown) were analysed in order to define the effects that the implementation is able to produce. In the following figures, these micro-activities were underlined inside process cycles.

OPEN IN VIEWER

Table 4.

Micro-activities of the case study

Ai Main Processes	Aik Macro-activities	Aijk Micro-activities
A2 Production	A21 Materials Entrance	A211 SAP transport order confirmation
		A212 SAP consumption confirmation
	A24 Products' Exit	A241 Insert weight on SAP system
		A241 Transport request creation
A8 Inventory	A81 Entrance	A811 Pallet identification
		A812 SAP transport order confirmation
	A82 Exit	A821 Pallet localization
		A822 Pallet identification
		A823 SAP transport order confirmation

3.2.1.1 Pallet creation

The pallet creation phase (see figure 3) is one phase which is less affected by the introduction of the technology: this is due to the fact that almost all the activities need to be manually performed or require sensitive data being entered into the managing system, even after RFID introduction. Indeed, in this phase the operator is called upon to enter the information regarding the material present on the pallet, to select the production line and production orders to which it will be associated, to create the pallet transport order on SAP and to print the relevant logistic fiche.

OPEN IN VIEWER

Figure 3.

Pallet creation processes

From Table 5 is clear that the introduction of RFID tags allows a process time reduction only for micro-activity A 241 ( 3 ) . On the other hand (see table 6), benefits in terms of ER and RCR are obtainable from the overall pallet creation macro-activity.

OPEN IN VIEWER

Table 5.

Pallet creation process savings

	u.m.	A (2) 241	A (2) 242
PTR	[s/u]	19.7	−2.3
γ	[%]	0.0	0.0

OPEN IN VIEWER

Table 6.

Pallet creation process savings (A⁽²⁾₂₄)

	u.m.	AS-IS	TO-BE	Value
error	[%]	4.30	0.80
T	[s/u]	123.5	106.1
ER	[s/u]	5.31	0.85	4.46
PI	[d]	–	–	–
RCR	[€]	–	3000	−3000

3.2.1.2 Pallets entrance in the coldstore

The pallet entrance in the coldstore phase (see figure 4) is certainly more affected by RFID introduction. The only unchanged operation is the pure manual pallet transfer with the forklift up to the cold room.

OPEN IN VIEWER

Figure 4.

Pallet entrance to the coldstore

The barcode scanning activity and transport order confirmation in SAP are the operations made automatic: as mentioned above, they become superfluous by the tag detection and subsequent information exchange with the informative system. This involves not only process time reduction for the forklift operator, as shown in table 7, but also other savings due to error and resource consumption reduction (see table 8).

OPEN IN VIEWER

Table 7.

Time savings of pallet entrance in the coldstore process

	u.m.	A (2) 811	A (2) 812
PTR	[s/u]	0.2	2.0
γ	[%]	100.0	0.0

OPEN IN VIEWER

Table 8.

Time savings of pallet entrance in the coldstore A⁽²⁾₈₁

	u.m.	AS-IS	TO-BE	Value
error	[%]	0.50	0.30
T	[s/u]	63.0	60.8
ER	[%/u]	0.32	0.18	0.14
PI	[d]	–	–	–
RCR	[€]	–	1000	−1000

In fact, the presence of a dedicated person to scanning and confirmation is no longer needed.

3.2.1.3 Pallets shipping from the coldstore

The pallet shipping from the coldstore phase (see figure 5) is analogous to the previous ones described above and in the same way is strongly impacted by the introduction of the RFID technology.

OPEN IN VIEWER

Figure 5.

Pallet shipping from the coldstore

The time needed for pallet transportation from the coldstore to the production halls and in the same manner the time needed by the operator to create the SAP transport order remain unchanged.

Most of the time saved in this phase (see table 9) is due to the pallet localization micro-activity that becomes automatic in the TO-BE process. Other important advantages for this phase are shown in table 10.

OPEN IN VIEWER

Table 9.

Time savings of pallet shipping from coldstore process

	u.m.	A (2) 821	A (2) 822	A (2) 823
PTR	[s/u]	67.8	0.2	2.0
γ	[%]	100.0	100.0	0.0

OPEN IN VIEWER

Table 10.

Time savings of pallet shipping from coldstore A⁽²⁾₈₂

	u.m.	AS-IS	TO-BE	Value
error	[%]	1.00	0.20
T	[s/u]	134.8	66.8
ER	[%/u]	1.35	0.13	1.22
PI	[d]	–	–	–
RCR	[€]	–	10000	−10000

3.2.1.4 Pallet arrival to the production system

The pallet arrival to the production halls and the material consumption in the production process phase (see figure 6), in terms of time, is strongly influenced by the introduction of the RFID technology. Upon arrival, the material is deposited in hoppers or in tilting machines in order to be able to start to feed the production process, and this operation remains unchanged in both configurations.

OPEN IN VIEWER

Figure 6.

Pallet arrival to the production system

The operation that leads to time savings (see table 11) is the automatic confirmation of the transport order, that in the AS-IS process is manually performed by an operator on the system. The tracking system performs these tasks automatically after detecting the RFID tags and reading the data. Savings in terms of ER and RCR are reported in table 12.

OPEN IN VIEWER

Table 11.

Time savings of pallet arrival to the production system

	u.m.	A (2) 211	A (2) 212
PTR	[s/u]	0.2	2.0
γ	[%]	100.0	0.0

OPEN IN VIEWER

Table 12.

Time savings of pallets arrival to the production A⁽²⁾₂₁

	u.m.	AS-IS	TO-BE	Value
error	[%]	0.50	0.30
T	[s/u]	63.0	60.8
ER	[%/u]	0.32	0.18	0.14
PI	[d]	–	–	–
RCR	[€]	–	1000	−1000

3.3. Economic assessment

In this section the performed cost benefit analysis is shown: in fact, to determine the true affordability of an investment, it is impossible to leave out economic considerations. In the previous sections the time savings for each of the four main logistic steps have been identified and quantified. For this analysis, it is worth distinguishing between tangible benefits, and therefore easily “monetizable”, and intangible ones, that can be achieved through the introduction of RFID technology, but that are more qualitative than quantitative.

3.3.1. Operational costs

Tangible benefits are evident: they have to be intended as a cost-cutting previously incurred or increased revenues not previously obtainable. In the presented business case, in the phases of pallet entrance and shipping to and from the coldstore, and of the arrival to the production, a dedicated person becomes unnecessary. This operator could be, for instance, reallocated to another process. In order to assess the savings associated to the reduction of process times, it is necessary to consider the number per year of moved pallets. From the figure ?? (not considering PH3, because this is reserved to the production of a single product and not included in the analysis as this not part of the pilot project) it is possible to derive the number of pallets/year passing through each phase and listed in the following table.

As it can be seen, other pallets used for production come from the raw material warehouse (at room temperature), but they are not considered. For the pilot project under consideration, in fact, only the phases mentioned above are taken into account.

In Table 13 and table 14, operational parameters for each macro-activity and operational global costs parameters are reported. Under the hypothesis, accepted by the company, that these resources might be absorbed by reallocating persons to other missions, it is concluded that the presented economic savings are realizable, see table 15.

OPEN IN VIEWER

Table 13.

Operational costs parameters for macro-activities

	u.m.	A (2) 24	A (2) 81	A (2) 82	A (2) 21
μ	[u/y]	50000	60000	60000	55000
σ	[u/y]	1000	1000	1000	1000
α	[%/y]	1.0	1.0	1.0	1.0
δ	[%/y]	3.0	3.0	3.0	3.0
CL	[€/h]	40000	40000	40000	40000
CI	[€/u]	400.0	400.0	400.0	400.0
CS	[€/u]	500.0	500.0	500.0	500.0
CM	[€/h]	2.20	2.20	2.20	2.20

OPEN IN VIEWER

Table 14.

Operational global costs parameters

	u.m.	Value
CTG	[€/y]	1.00
CHD*	[€]	118560
β	[%/y]	2.0
FTE	[h/y]	1496

*

see paragraph 3.3.2.

OPEN IN VIEWER

Table 15.

Operational costs of macro-activities (€)

	A (2) 24	A (2) 81	A (2) 82	A (2) 21	Total
ΔCL	7953.8	1035.6	31553.7	949.3	41492.4
ΔCI	79.5	10.4	315.5	9.49	414.9
ΔCS	99.4	12.9	394.4	11.9	518.6
ΔCM	−590.0	−622.7	1593.3	−573.3	−192.6
CHD · β					2371.2
ΔCT					39862.1

OPEN IN VIEWER

Figure 7.

Pallet per year flow among the warehouse, the production halls and the coldstore

The RFID technology introduction in the logistic process has not only led to the reduction of execution times of the processes reported in the precedent paragraphs. Hereafter some of the other benefits deriving from the application of this technology will be illustrated. They are the so-called intangible benefits important for the objectives of error reduction and production streamlining.

3.3.2. Implementation costs

A fundamental part of the pilot project has also been cost evaluation and the choice of the supplier would then help the company and support the future development of this technology in its business. Table 16 shows the hardware components of the RFID system, the quantities and costs. In addition, we indicate the total cost for the purchase of licenses and for the integration of the RFID system with enterprise management software.

OPEN IN VIEWER

Table 16.

Implementation costs

	#	€	Total
Tag	60000	1.00	60000
Ant. stair	4	4320	17280
Ant. PH2	2	2320	4640
Ant. PH3	2	2320	4640
Middleware soft.	1	32000	32000
CHD			118560

3.3.3. Profitability analysis

In this paragraph the conclusions on the previously shown financial indexes are taken. In fact, the profitability of the investment is well understood looking at the yearly cost parameters. The NPV has been calculated considering an investment duration of five years and a discount rate of 5.0% per year. The results of all profitability indicators are reported in table 17.

OPEN IN VIEWER

Table 17.

Profitability indicators

	u.m.	Value
NPV	[€]	62651
IRR	[%]	35.7
PBP	[y]	3.2

3.3.4. Sensitivity and risk analysis

In Table 18 we illustrate the parameter values that influence the NPV. The value of these parameters has been increased by 20% and the new NPVs (+20%) have been calculated for each increment. The percentage increase in NPV (Δ(%)) from baseline (€62 651) and the most significant parameters (p*) have been identified.

OPEN IN VIEWER

Table 18.

Sensitivity analysis

		Value		NPV
#	Parameter	Base	+20%	+20%	Δ(%)	p*
1	PTR(A241)	19.0	23.64	69452	10.86	*
2	PTR(A242)	−2.3	−2.76	61857	−1.27
3	PTR(A811)	0.2	0.24	62740	0.14
4	PTR(A812)	2.0	2.40	63479	1.32
5	PTR(A821)	67.8	81.36	92999	48.44	*
6	PTR(A822)	0.2	0.24	62740	0.14
7	PTR(A823)	2.0	2.4	63479	1.32
8	PTR(A211)	0.2	0.24	62733	0.13
9	PTR(A212)	2.0	2.4	63410	1.21
10	error(A24) (to-be)	0.80	0.96	62334	−0.51
11	error(A81) (to-be)	0.30	0.36	62574	−0.12
12	error(A82) (to-be)	0.20	0.24	62567	−0.13
13	error(A21) (to-be)	0.30	0.36	62580	−0.11
14	T(A24) (to-be)	106.1	127.32	63292	1.02
15	T(A81) (to-be)	60.8	73.00	62676	0.04
16	T(A82) (to-be)	66.8	80.16	62762	0.18
17	T(A21) (to-be)	60.8	73.00	62674	0.04
18	μ(A24)	50000	60000	69591	11.08	*
19	μ(A81)	60000	72000	49984	−20.22	*
20	μ(A82)	60000	72000	80615	28.67	*
21	μ(A21)	55000	66000	56494	−9.83	*
22	σ(A24)	1000	1200	62651	0.00
23	σ(A81)	1000	1200	62651	0.00
24	σ(A82)	1000	1200	62651	0.00
25	σ(A21)	1000	1200	62651	0.00
26	α	1.00	1.20	60605	−3.27
27	δ	3.00	3.60	62242	−0.65
28	CL	40000	48000	100375	60.21	*
29	CI	400	480	63028	0.60
30	CS	500	600	63122	0.75
31	CM	2.20	2.64	64931	3.64	*
32	CTG	1.00	1.20	47515	−24.16	*
33	β	2.0	2.4	60495	3.44
34	Discount rate (i)	5.0	6.0	59428	5.14	*

Pessimistic (PE), Optimistic (OE) and Most Likeable (ML) values and the event probabilities have been estimated for this parameters (see table 19).

OPEN IN VIEWER

Table 19.

Sensitivity analysis

	Value			Probability
Parameter	PE	ML	OE	%PE	%ML	%OE
PTR(A241)	25.0	19.7	15.0	20	70	10
PTR(A821)	80.0	67.8	55.0	20	70	10
μ(A24)	48000	50000	55000	20	70	10
μ(A81)	50000	60000	65000	30	60	10
μ(A82)	50000	60000	65000	30	60	10
μ(A21)	50000	55000	65000	30	60	10
CL	45000	40000	38000	10	70	20
CM	2.50	2.20	2.00	15	70	15
CTG	1.50	1.00	0.80	15	70	15
Discount rate (i)	7.0	5.0	4.0	30	60	10

Calculating NPV value for all 59 049 (3¹⁰) factor combinations and the event probability value related to each combination and putting NPV values in increasing order, it is possible to draw the curve in figure 8. From this curve we can identify a 3.1% investment risk, that is the probability that NPV is less than zero.