=Paper= {{Paper |id=Vol-1420/ilog-paper6 |storemode=property |title=Improvements of Logistics in Region Campania Using a Profiling/Competence-based Approach, Enriched with Experience |pdfUrl=https://ceur-ws.org/Vol-1420/ilog-paper6.pdf |volume=Vol-1420 |dblpUrl=https://dblp.org/rec/conf/bis/BassanoCDGR15 }} ==Improvements of Logistics in Region Campania Using a Profiling/Competence-based Approach, Enriched with Experience== https://ceur-ws.org/Vol-1420/ilog-paper6.pdf

Improvements of Logistics in Region Campania Using a
Profiling/Competence-based Approach, Enriched with
Experience

Clara Bassano1, Maria Vincenza Ciasullo2, Giuseppe D’Aniello3,
Matteo Gaeta3, and Luigi Rarità3
1
Dipartimento di Studi Aziendali e Quantitativi,
University Parthenope of Naples,
Via Generale Parisi, 13 (Palazzo Pakanowski), 80132, Naples, Italy
clara.bassano@uniparthenope.it
2
Dipartimento di Studi e Ricerche Aziendali – Management & Information Technology,
University of Salerno, Via Giovanni Paolo II, 132, 84084, Fisciano (SA), Italy,
mciasullo@unisa.it
2
Dipartimento di Ingegneria dell’Informazione, Ingegneria Elettrica e Matematica Applicata,
University of Salerno, Via Giovanni Paolo II, 132, 84084, Fisciano (SA), Italy
{gidaniello, mgaeta, lrarita}@unisa.it

Abstract. This paper focuses on improvements of logistics via removal of
weakness points inside the district Agro Nocerino Sarnese (Italy, region Cam-
pania), famous for the export of products of high quality to other Italian regions
and/or foreign countries. The logistic chain is guaranteed by fleets of trucks,
whose aim is to transport goods from a point to another inside the region. In or-
der to repair eventual faults of trucks, the logistic network has particular nodes
that are small family business, which have decision processes of naïf type deal-
ing with tradition and related to a leadership. In absence of such leadership, a
management crisis might occur, with consequent delays on repair processes and
on the delivery of products to the final customers. Hence, possible improve-
ments on performances of the logistic network are achieved trying to reproduce
the leadership’s decisions (making them stable) via a profiling/competence-
based approach, enriched by experience. Precisely, models, dealing with pro-
files of clients, competences of workers and experience of the leadership, are
mixed up and used to reconstruct the leadership’s decisions. A real case of
small business is useful to test the proposed approach.

Keywords: logistics, competence, profiling, experience

1 Introduction and motivations

Phenomena connected to logistics inside Italian regions are nowadays object of
particular attention, due to the survival of regional economy as well as the wealth of
the Italy itself. Indeed, region Campania represents a geographical portion in which
some aspects connected to transports and distribution of products are fundamental in

Copyright © 2015 by the authors. Copying permitted for private and academic purposes.
This volume is published and copyrighted by its editors.

74
order to guarantee profits. This is evident thinking of various goods that region
Campania produces. For instance, the famous pizza “di Napoli” is obtained via some
operations that derive from high quality products, such as the tomatoes of San
Marzano and/or the water of sources near the Vesuvius. It is obvious that other parts
of Italy, as well as other foreign countries, require products of region Campania, with
the aim of reproducing foods and meals of acceptable standards.
On the other hand, region Campania has the necessity of exporting its own
products, as the primary intention is an increment of prestige at national and
international level. For this reason, various efforts are made in order to guarantee
efficient supply chains, where the satisfaction of producers meets the one of
consumers, also thanks to an efficient logistic network that involves the whole region
Campania. In particular, the so-called district “Agro Nocerino Sarnese” (ANS),
namely an area in Campania, is useful to understand the complex dynamics of
region’s logistics. In fact, the district ANS is characterized by the production of many
consumer goods, that are then distributed throughout the region and exported to other
parts of Italy and/or to foreign countries. One factor that implies wealth in region
Campania is the quality of goods that are produced by the district. In particular, the
production and the consequent export are the key factor for the success of all
operations, which occur inside the region.
In particular, the transport of products from a point to another is often guaranteed
by “padroncini”, i.e. owners of trucks. Padroncini aim at managing their fleets of
vehicles in order to achieve high quality standards in terms of relationships of type
producer-consumer. Indeed, extraordinary events, such as malfunctions and technical
problems of trucks, might cause failures for the products’ transports, with consequent
disadvantages for the final customer and the performances of the overall logistic net-
work of region Campania.
Indeed, the logistic network has some particular nodes, consisting of small family
business, whose job is to repair possible faults of trucks. It follows that a possible
high critical situation for logistics inside region Campania depends on how activities
of small family business are managed, see [9]. In fact, delays in repairing faults are
easily translated into delays in distribution and consequent deterioration of products.
As a consequence, nodes, which represent small family businesses, influence the
logistic chain. Hence, a possible policy for improvements foresees the reproduction of
leadership’s decisions for the work teams assigned at each work order. In fact, the
leadership, thanks to its experience, knows the client that has to be served first and
how to serve him, see [4]. Indeed, such problem is very complex, as decision criteria
are based on experience and tradition of the company and the leadership might vary
during years. This last aspect represents a fundamental point, as wrong decisions
could cause negative consequences on the served customers (in our case, padroncini).
In this paper, with the aim of improving the logistics inside region Campania, we
study a possible methodology for capturing the leadership’s decisions of family
businesses. Such approach foresees the fusion of models that are often used
individually or in pipe and deal with profiles, competences and experience, i.e. it is a
profiling/competence-based approach, shaded with experience.

75
In particular, as for profiles, the clients’ characteristics are studied in order to
establish some priorities among the work orders that arrive at a generic small
business, see [5], [10]; as for competences, a model based on Knowledge, Skills and
Attitude is considered (for further explainations, see [6], [7], [8], [17], [18] and [20]);
finally, as for experience, a Pattern Mining algorithm, and precisely an Apriori one
(more details are in [1], [2], [3], [11], [12], [13], [14], [15], [16], [19], [21]), is used to
understand if the leadership proposes typical group of workers. This last aspect is
very interesting because the Apriori algorithm is useful to understand all those
features that are difficult to reproduce with the usual knowledge methodologies.
The approach is tested on a real small business inside region Campania. The results
for work teams of first choice (the teams chosen by the leadership) and second choice
(the alternative teams if first choices are not possible) are very positive. In particular,
for the work teams of second choice, the proposed approach indicates that 90% of
leadership’s decision is reproduced (also if the leadership is absent, and this makes
the process stable).
The paper is organized as follows. In Section 2 the real conditions of logistics
inside region Campania are described, with emphasis on the transport of goods and
problems due to unexpected events. Section 3 considers the case study of a real
critical node in region Campania. Section 4 reports the final research results. The
paper ends with Conclusions in Section 5.

2 Logistics in region Campania

Inside region Campania, Italy, there is the area of Agro Nocerino – Sarnese (usual-
ly called district ANS), which consists of two different geographical parts, the Agro
Sarnese and the Agro Nocerino, both located in the valley of the river Sarno, precisely
between Naples and Salerno. The district ANS produces many consumer goods, dis-
tributed throughout the region and exported to other parts of Italy. The main groups of
these products are the following: meat, fish, fruit, milk and dairy products, vegetables
(in particular, tomatoes and onions).
Obviously, depending on the type, the previous goods undergo several steps in the
producer-consumer chain. Their transit to the consumer is due to a dense logistic net-
work, which involves the whole region Campania. In this context, many factors allow
the success of the operations. In particular, we consider the speed of transportation
from one point to another in the region, the need to maintain appropriate conditions
on the goods during transports (for example, the existence of the cold chain for fish),
the partial reduction of unexpected events, such as faults to the means that allow
transits in the various points of the network. Notice that, as for big distances and
quantities, large transport companies with internal workshops manage trucks that
provide the location of goods to another node of the logistics network, thus ensuring
an effective, efficient and redundant service. On the contrary, the whole universe of
small and medium retailers, served primarily by “padroncini”, deals with the regional
network.

76
These small transportation companies, often consisting of just one person, are fun-
damental to ensure the supply of products to small and medium-sized municipalities
as they represent the backbone of the region. Padroncini use a number of work-
shops/establishments, which are inside the district ANS and are specialized in repair,
customization and maintenance of trucks. An example of workshop is the Santonicola
F.lli SNC, located in Siano, province of Salerno.
The establishments manage the padroncini’s demands because means useful for the
transports of goods are often subject to malfunctions of different types, such as me-
chanical failures for the transport itself and technical problems relating to the storage
of the product. The latter is much more critical because the necessary condition that
guarantees intact products inside distribution centres is that conservation devices of
trucks could work in optimal conditions. Considering such phenomena, as for the
transits of trucks in the logistic network, there are further nodes, consisting of sets of
small family businesses, whose task is to repair eventual faults of trucks. This sug-
gests that a possible optimization or, in better terms, a possible improvement of the
internal dynamics of the district ANS may depend on the management of activities
inside small family business. Such situation is easy to understand if you think that
delays in repairs of faults imply cascade delays for the distribution of products inside
distribution centres, with consequent deterioration of products before they arrive at
the final consumer. Hence, nodes representing small business are critical, as they
influence the logistic distribution inside region Campania.
In order to overcome such problems, it is necessary to focus on these nodes and
their activities. This subject is complex because the decision criteria within small
businesses are very empirical, based on experience and tradition and taken by a lead-
ership that, over the years, may be subject to changes. The transition from a leader-
ship to the next one may cause negative trends, as possible decisions are not struc-
tured and the whole knowledge of the previous leadership is not transferred to the
new generation. This has meaningful impacts on either served customers (in this case,
padroncini) or the company itself. Such situations imply, undoubtedly, negative per-
formances inside the logistic chain.
Considering the scenario of the district ANS, the aim of this analysis is to study a
possible critical node, providing for it an appropriate solution to address the needs of
padroncini and ensuring, therefore, a possible improvement of performances inside
the logistic network.

3 Analysis of a real critical node

In what follows, we consider the analysis of a real critical node of the logistic
network inside region Campania. The activities of such enterprise are managed by a
family and deal with single work orders that are satisfied according to clients’
exigencies.
In particular, the leadership considers each work order and, according to the
clients’ types and priority reasons, assignes a work team, which consists of a subset of
workers, the most suitable either for their skills/abilities or for the characteristics of

77
work orders and priorities of padroncini. The enterprise has different workers, who
are listed as follows and whose names are not reported for privacy. There are four
coach builders (C1, C2, C3 and C4), four varnishers (V1, V2, V3 and V4), four
welders (W1, W2, W3 and W4), two electricians (E1 and E2), three mechanics (M1,
M2 and M3) and two upholsterers (U1 and U2).
We consider sixteen work orders to which the leadership assigned a work team.
Unlike the case reported in [9], a deeper analysis allowed establishing the so-called
teams of “second choice”, i. e. the teams chosen by the leadership when some of the
most suitable workers for a given work order are already busy or not available. The
analysed work orders and the work teams of second choice are in Table 1.

Table 1. Work orders and the corresponding work teams of second choice

Work order Work team Work order Work team
1 – Engine restoration C1, M3, V1 9 – Leaf spring substitution W4
2 – Mobile case restoration V2, W1 10 – Cabin substitution C2, E1, M3, V3
3 – Refrigerator restoration V2 11 – Truck recovery C2, E2, M2, V1
4 – Cargo bed modifications V1, W4 12 – Truck restoration E1, M1, V2, W1
C2, E2, U2,
5 – Truck preparation V2, V3, W1, W3 13 – Tractor transformation
V1, V2
6 – Pitch stretching E1, M1, V3, W3 14 – Soft top construction E1, V4, W2, W3
7 – Case painting V1, W4 15 – Cistern painting C3, C4, V1, V3
8 – Cabin painting C2, V1 16 – Bartolini case restoration V3, V4, W3, W4

Notice that the work teams of second choice are not always used by the leadership.
Indeed, for some particular work orders (1, 9, 12, 14, 15 and 16, indicated in gray in
Table 1), the enterprise does not prefer to assign alternative teams because of the
delicacy of the required actions. In these cases, the work teams are the first choices of
the leadership. Hence, if it is not possible to assign the team of “first choice”, the
work order is queued and the corresponding operations are made in a second moment.
Moreover, queues of work orders might also occur if it is not possible to create work
teams of second choice. In this last case, the work orders are completed with a
priority, which is strictly dependent on the clients’ characteristics.
In order to improve all operations inside the logistic network, an obvious analysis
of work orders and teams is necessary, with the aim of reproducing the possible
choices of the leadership. This problem is very difficult, as it deals with the tradition
and history of the enterprise. For this reason, various approaches, whose combination
might imply a partial reproduction of the leadership’s choices, should be considered.
Decomposing the original problem into possible factors that are suitable to find a
correct solution, it is possible to recognize that the choices’ characteristics obey the
following criteria: client’s features; competences of workers involved in work teams;
leadership’s experience. Hence, the approach that we are going to follow is
profiling/competence – based and it is enriched with experience. Indeed, there are
many models dealing with such problems, but are often used separately or in pipe.

78
The real efforts in this work foresee the fusion of existing approaches, with the aim of
obtaining an acceptable reconstruction of leadership’s decisions.
In detail, as for the profiling, we follow a model based on convex combinations of
factors (see [10] for an example) that are useful to represent the enterprise’s clients.
As for the competence representation, we consider Knowledge, Skills and Atti-
tudes, see [18] for details. In particular, Knowledge is the set of support information
for a given task; Skill is the capacity of developing the task; Attitude indicates a par-
ticular behavior in facing some situations. Indeed, we indicate as KSA Model (see [7],
[8], [17], [20]) the competence representation in terms of Knowledge (K), Skills (S)
and Attitudes (A). Such model is implemented via Lightweight Ontologies (written in
SKOS and similar to taxonomies, see [6]), whose aim is to model a particular domain
in a hierarchical way. In particular, each element of type K, S and A has a score, that
indicates the competence levels for a particular knowledge domain.
Finally, as for the leadership’s experience, Pattern Mining techniques are useful to
find relevant patterns in data sequences. In this paper, we use a Pattern Mining algo-
rithm called “Apriori” (see [1], [2], [9], [11], [12], [14], [15], [16], [19], [21]).
In what follows, first we describe some features of the chosen methods for
profiling, competences and experience; then, we briefly analyse the approach for the
reconstruction of leadership’s decisions.

3.1 Profile modelling
Suppose that each client Ci , i = 1,..., KC , of the enterprise is indicated by a vector,
whose components are the following:
• Sales figures, Ci1 .
• Time payments, Ci 2 .
• Brand, Ci 3 , seen as a measure of the company’s perception.
• Number of annual reports, Ci 4 .
• Loyalty, Ci 5 .
For the client Ci , i = 1,..., KC , the component Ci1 is given by the average sale fig-
ures in last three years, while the component Ci 3 indicates the average number of
work orders in last three years. The remaining parameters Ci 2 , Ci 4 and Ci 5 are varia-
ble and generally estimated by the enterprise.
The priority of each client Ci , i = 1,..., KC , is computed as:

5
pCi = ∑ ω j Ci j ,
j =1

5
where 0 < ω j < 1, j = 1,...,5, ∑ ω = 1.j
j =1

79
3.2 Competence modelling

For the analysed enterprise, two different types of KSA models are considered,
precisely for workers and work orders. Notice that a different KSA model is defined
for each type of worker (coach builder, electrician, mechanic, upholsterer, varnisher
and welder). Then, following the characteristics of the various work orders, KSA
models for work orders are constructed. Indeed, KSA models for workers and work
orders have a precise difference: the former focus on all characteristics of workers
while the latter consider what is useful for an assigned work order. For a better
comprehension, consider Table 2, that reports the thirtytwo concepts of the KSA
model for the worker “varnisher”, divided into Knowledge, Skills and Attitudes. In
particular, the eight bold concepts allow constructing the KSA model for the work
order 3.

Table 2. Complete KSA model for the worker “varnisher”
Knowledge Skills Attitudes
Reading
Abrasive Application of safety Quality check for
instructions Accuracy
materials procedures in production the done work
on manuals
Application of
Application of routine
Thinners and Preparation of painting
maintenance for plants Manual skills
solvents surfaces to paint techniques on
and equipment
metals
Protection of Application of
Running the painting by
the area procedures for Flexibility and
Regulations sprinklers inside booths
surrounding the maintenance and adaptability
for painting
objects to paint plants machinery
Reaction of Application of
Preparation of
materials Application of further procedures for Work in team
the spray
in painting paintworks noncompliance of and cooperation
equipment
treatments unfinished parts
Adjustment of
Application of
the equipment
Features of Application of any techniques to
according to the
paints other materials clean metal
features of
surfaces
materials
Application of Transport of painted Use of
Mechanical
criteria for paint parts in ovens and instruments for
drawing
preparation interest for drying stages painting
Visual check or
Specifications Use of personal Application of
measurement of the
of metal protection quality control
thickness of the paint
materials devices procedures
application

Notice that the concepts for each KSA model (either for workers or for work
orders) correspond to different skos:ConceptScheme and skos:Concept, as shown in
Fig. 1, that represents an extract of the KSA Model for Welders (shortly indicated by
KMW).

80
Fig. 1. A portion of KMW

From Fig. 1, we get that the category Knowledge has the subcategories Mechanics
and Electronics; Skills has treatment of materials and use of devices, and so on. The
ellipses indicate the subsets of welder qualities useful to construct a Worker Order
KSA model (shortly indicated by WOKMW), and precisely the work order 9. Notice
that, using the only KMW, the welder to choose is W2 (the best worker for the work
order), while WOKMW proposes W4. This last choice is coherent with the leadership
one (see Table 1), as it indicates the best worker for the assigned work order.

3.3 Shades of experience
The interaction among workers is an important parameter for the success of an en-
terprise. Indeed, the leadership often foresees that some workers have a high degree of
cooperation. Such phenomenon is difficult to model, as it deals with the behavioural
attitudes of single workers and the empathy with other colleagues. This situation is
captured via the “Apriori” algorithm. It considers that, if a given item set is frequent,
its subsets are frequent too. Precisely, consider a transaction database T and a support
threshold !. Let !! be the candidate item set of length k and !! be the frequent item
set of length k. A possible pseudo code for the algorithm is the following:

Apriori (T , ε )
L1 ← {itemset of length 1}
k ←2
while Lk −1 ≠ ∅
Ck ← generate Lk −1
for transactions t ∈ T , Ct ← subset (Ck , t ) ;

for candidates c ∈ Ct , count [c] ← count [c] + 1; Lk ← {c ∈ Ck : count [c ] ≥ ε } ;

81
k ← k +1
return U Lk
k

The Apriori algorithm was able to identify the following couples of workers that
often occur in work teams of second choices: (C2, V1), (V1, W2), (V3, W3), (E1,
M1) and (V1, W4). Such couples represent persons who often work together, and
depend only on the leadership itself, i.e. on its experience. The found couples are a
quite meaningful example to capture the leadership’s experience, a heritage that is
difficult to transmit to future generations.

3.4 The whole approach
The whole approach foresees the fusion of the three chosen models (often used in-
dividually, see subsections 3.1, 3.2 and 3.3) in order to capture suitable work teams of
first and second choice. The steps of the approach are the following (consider Fig. 2
for an abstract vision):
• Reorder the work orders according to the clients’ importance and priori-
ties.
• For each work order, choose the team obtained via the KSA models. If
such teams coincides with some choices of the leadership, stop; otherwise,
try to change workers of work teams using the association rules obtained
by the Apriori algorithm.
• If, for the work order in consideration, first and second choice’s teams
have already been assigned, put the client in queue.

Fig. 2. Chain of the proposed approach

4 Results

In this section, we present the results obtained for the enterprise described in Sec-
tion 3. In particular, Table 3 reports the comparison between the leadership’s work
teams (LWT) of first and second choice for the various work orders (WOs) and the
ones obtained via the fusion among profiling modelling (clients’ characteristics),
competence modelling (KSA models), and experience (APriori algorithm). Gray lines

82
indicate cases in which the leadership does not foresee second choices, and the recon-
struction’s percentage is also indicated for each reconstructed team.

Table 3. Work orders of first and second choice and their reconstruction
LWT Reconstructed LWT Reconstructed
WOs
of first choice teams of first choice of second choice teams of second choice

1 C1, M3, V1 C1, M3, V1 (100%) C1, M3, V1 Not applied

2 V3, W2 V3, W2 (100%) V2, W1 V2, W1 (100 %)

3 V4 V4 (100%) V2 V2 (100 %)

4 V4, W4 V4, W4 (100%) V1, W4 V1, W4 (100 %)

5 V1, V2, W2, W3 V1, V3, W2, W3 (75%) V2, V3, W1, W3 V2, V3, W1, W3 (100 %)

6 E1, M1, V3, W2 E1, M1, V3, W2 (100%) E1, M1, V3, W3 E1, M1, V3, W2 (100 %)

7 V4, W4 V4, W4 (100 %) V1, W4 V1, W4 (100 %)

8 C2, V1 C1, V1 (100%) C2, V1 C2, V1 (100%)

9 W4 W4 (100%) W4 Not applied

10 C2, E1, M3, V2 C1, E1, M3, V3 (50%) C2, E1, M1, V3 C2, E1, M1, V3 (50 %)

11 C1, E2, M2, V1 C1, E2, M3, V1 (75%) C2, E2, M2, V1 C2, E2, M1, V1 (75 %)

12 E1, M1, V2, W1 E1, M1, V3, W2 (50%) E1, M1, V2, W1 Not applied

13 C1, E2, U2, V1, V3 C1, E2, U1, V1, V3 (80%) C2, E2, U2, V1, V2 C2, E2, U2, V1, V2 (100 %)

14 E1, V4, W2, W3 E2, V3, W2, W3 (50%) E1, V4, W2, W3 Not applied

15 C3, C4, V1, V3 C1, C4, V1, V3 (75%) C3, C4, V1, V3 Not applied

16 V3, V4, W3, W4 V3, V4, W3, W4 (100%) V3, V4, W3, W4 Not applied

As for the first choices, the approach is able to capture a total correspondence for
the work orders 1, 2, 3, 4, 6, 7, 8, 9 and 16, i.e. a 100% reconstruction occur for 9/16
situations. As for second choices, considering that work orders 1, 9, 12, 14, 15 and 16
have not to be analyzed, we have a 100% correspondence in cases 2, 3, 4, 5, 6, 7, 8,
11 and 13.
Such results are in Table 4 in terms of correspondence percentages.

Table 4. Correspondence percentages for work teams of first and second choice

Case 100% More than 50% Less or equal than 50%
Work teams of first choice 9/16 13/16 3/16
Work teams of second choice 8/10 9/10 1/10

Notice that, using the fusion of the various approaches described in Section 3, we
have that:
• For the work teams of first choice, 13/16 work orders (about 80%) present
a more than 50% correspondence with the leadership’s decisions.

83
• For the work teams of second choice, 9/10 work orders (90%) have a
more than 50% correspondence with the leadership’s decisions.
This indicates that the proposed approach is quite useful for the reconstruction of
the enterprise’s activity with a high degree of accuracy.

10
8
6
4
2
0
100% More than 50 % Less or equal
than 50%

Fig. 3. Number of work teams of second choice with more than 50 % correspondence

5 Conclusions

In this paper, an analysis of logistics inside the district Agro Nocerino Sarnese (Ita-
ly, region Campania) is studied. Possible improvements are obtained via the reproduc-
tion of decisions inside small family businesses.
Considering a real case of enterprise and using a profiling/competence-based ap-
proach, enriched with experience, it was possible to reproduce the dynamics of lead-
ership’s choices with a 80% and 90% degree accuracy, respectively, for work teams
of first and second choice.

6 References
1. Agrawal, R., Srikant, R.: Fast Algorithms for Mining Association Rules, In: Proceedings
of the 20th VLDB Conference, pp. 487 – 493 (1994)
2. Ashtana, P., Singh, A., Singh, D.: A Survey on Association Rule Mining Using Apriori
Based Algorithm and Hash Based Methods. International Journal of Advanced Research in
Computer Science and Software Engineering. 3(7), pp. 599 – 603 (2013)
3. Barile, S., Saviano M., Polese F.: Information asymmetry and co-creation in health care
services. Australian Marketing Journal. 22(3), pp. 205 – 217 (2014)
4. Cabrera-Suárez, K.: Leadership transfer and the successor’s development in the family
ﬁrm. The Leadership Quarterly. 16, pp. 71 – 96 (2005)
5. Capuano, N., D’Aniello, G., Gaeta, A., Miranda, S.: A personality based adaptive
approach for information system. Computers in Human Behaviour. 44, pp. 156 – 165
(2015)
6. Clark, D.: Bloom’s Taxonomy of Learning Domains. Online version:
http://www.nwlink.com/~donclark/hrd/bloom.html
7. Del Nostro, P., Orciuoli, F., Paolozzi, S., Ritrovato, P., Toti, D.: A Semantic – Based
Architecture for Managing Knowledge – Intensive Organizations: The ARISTOTELE

84
Platform, In: Web Information Systems Engineering – WISE 2011 and 2012 Workshops,
pp. 133 – 146 (2013), Springer
8. Gaeta, A., Gaeta, M., Piciocchi, P., Ritrovato, P., Vollero, A.: Evaluation of the human
resources relevance in organisations via knowledge technologies and semantic social
network analysis. Internation Journal of Knowledge and Learning. In press (2014)
9. Gaeta, M., Perano, M., Piciocchi, P., Rarità, L.: Construction of decision criteria for family
– run business, In: Proceedings itAIS 2014, pp. 1 – 12 (2014)
10. Gaeta M., Piscopo R., Rarità L., Trevisant L., Novi D.: A Knowledge Management
Strategy to Identify an Expert in Enterprise. Smart Organizations and Smart Artifacts,
Lecture Notes in Information Systems and Organisation, Springer International Publishing
Switzerland 2014. 7, pp. 173 – 182 (2014)
11. Gu, J., Wang, B., Zhang, F., Wang, W., Gao, M.: An Improved Apriori Algorithm. Com-
munications in Computer and Information Science. 224, pp. 127 – 133 (2011)
12. Hong, T., Kuo, C., Wang, S.: A fuzzy AprioriTid mining algorithm with reduced
computational time. Applied Soft Computing. 5(1), pp. 1 – 10 (2004)
13. Hong, T. P., Wang C. Y., Lin C. W.: Providing timely updated sequential patterns in
decision making. International Journal of Information Technology & Decision Making.
9(6), pp. 873 – 888 (2010)
14. Jia, C. Y., Ni, X. J.: Association rule mining: A Survey. 30, pp. 145 – 149 (2011)
15. Kaur, C.: Association Rule Mining using Apriori Algorithm: A Survey. International
Journal of Advanced Research in Computer Engineering & Technology. 2(6), pp. 2081 –
2084 (2013)
16. Lin, C., Hong, T., Lan, G., Wong, J., Lin, W.: Incrementally mining high utility patterns
based on pre – large concept. Applied Intelligence. 40(2), pp. 343 – 357 (2014)
17. Loia, V.: Special issue on new trends for ontology – based knowledge discovery.
International Journal of Intelligent Systems. 25(12), pp. 1141 – 1142 (2010)
18. Loia, V., De Maio, C., Fenza, G., Orciuoli, F., Senatore, S.: An enhanced approach to
improve enterprise competency management, In: 2010 IEEE World Congress on
Computational Intelligence, WCCI (2010)
19. Patel, B., Chaudhari, V. K., Karan R., K., Rana YK: Optimization of Association Rule
Mining Apriori Algorithm Using ACO. International Journal of Soft Computing and
Engineering. 1(1), pp. 24 – 26 (2011)
20. Sampson, D., Fytros, D.: Competence models in technology – enhanced competence based
learning, In: H. H. Adelsberger, Kinshuk, J. M. Pawlowski, D. G. Sampson (Eds.),
Handbook on Information Technologies for Education and Training, International
Handbooks on Information Systems, Springer Berlin Heidelberg, pp. 155 – 167 (2008),
doi:10.1007/978-3-540-74155-8-9
21. Sumithra, R., Paul, S.: Using distributed apriori association rule and classical apriori
mining algorithms for grid based knowledge discovery, In: 2010 International Conference
on Computing Communication and Networking Technologies, pp. 1 – 5 (2010)