Evidence-based medical equipment

management: a convenient implementation

Ernesto Iadanza, Valentina Gonnelli,

Francesca Satta & Monica Gherardelli

Medical & Biological Engineering &

Computing

ISSN 0140-0118

Med Biol Eng Comput

DOI 10.1007/s11517-019-02021-x

1 23
Your article is published under the Creative
Commons Attribution license which allows
users to read, copy, distribute and make
derivative works, as long as the author of
the original work is cited. You may self-
archive this article on your own website, an
institutional repository or funder’s repository
and make it publicly available immediately.

1 23
Medical & Biological Engineering & Computing
https://doi.org/10.1007/s11517-019-02021-x

ORIGINAL ARTICLE

Evidence-based medical equipment management: a convenient

implementation
Ernesto Iadanza 1 & Valentina Gonnelli 1 & Francesca Satta 2 & Monica Gherardelli 1

Received: 5 July 2018 / Accepted: 27 July 2019

# The Author(s) 2019

Abstract
Maintenance is a crucial subject in medical equipment life cycle management. Evidence-based maintenance consists of the con-
tinuous performance monitoring of equipment, starting from the evidence—the current state in terms of failure history—and
improvement of its effectiveness by making the required changes. This process is very important for optimizing the use and
allocation of the available resources by clinical engineering departments. Medical equipment maintenance is composed of two
basic activities: scheduled maintenance and corrective maintenance. Both are needed for the management of the entire set of medical
equipment in a hospital. Because the classification of maintenance service work orders reveals specific issues related to frequent
problems and failures, specific codes have been applied to classify the corrective and scheduled maintenance work orders at Careggi
University Hospital (Florence, Italy). In this study, a novel set of key performance indicators is also proposed for evaluating medical
equipment maintenance performance. The purpose of this research is to combine these two evidence-based methods to assess every
aspect of the maintenance process and provide an objective and standardized approach that will support and enhance clinical
engineering activities. Starting from the evidence (i.e. failures), the results show that the combination of these two methods can
provide a periodical cross-analysis of maintenance performance that indicates the most appropriate procedures.

Keywords Evidence-based maintenance . Health technology management . Key performance indicators . Medical equipment .
Clinical engineering

1 Introduction is a key process throughout the life cycle of every medical

device. Maintenance planning requires the assessment of a
Today’s rapid and continuous technological evolution, which number of parameters, including how a piece of equipment
affects most production sectors, also involves healthcare. is used, how often it is used, its intended use, risk associated
Indeed, healthcare technologies have become an essential part with its usage and its failure rates.
of the provided services, as they play increasingly significant There are two main types of maintenance required for med-
roles in the diagnosis and treatment of patients. ical equipment in all hospitals: scheduled maintenance (SM)
The complexity of the technological assets found in and corrective maintenance (CM). SM, in compliance with the
healthcare facilities, in terms of number and diversity, is manufacturer’s instructions, includes the operations per-
reflected in the complexity of technology management, which formed at scheduled times to reduce deterioration from use
must be efficient so that the equipment can always be used (often referred to as “preventive maintenance”) or the occur-
safely and appropriately. From this perspective, maintenance rence of functional failures. CM comprises the repair of the
equipment’s functions (i.e. its restoration) as well as its re-
placement when repair is not feasible due to costs or obsoles-
* Ernesto Iadanza
[email protected] cence [15].
Maintenance is also a crucial aspect of the activities in a
1
hospital’s clinical engineering (CE) department because it in-
Information Engineering Department, University of Florence, Via S.
Marta, 3, 50139 Florence, Italy
volves significant human and financial resources. Therefore,
2
the assessment of the effectiveness of any maintenance
ESTAR - Dipartimento Tecnologie Informatiche e Sanitarie UOC,
Tecnologie Sanitarie AOU Careggi/Meyer, Largo Brambilla 3,
programmes is strictly linked to the optimization of the use
50141 Florence, Italy of available resources in CE departments [20].
 Med Biol Eng Comput

This is the context of this research work, which shows an Regionale), which involves 6 engineers, 5 technicians, 2 ad-
evidence-based approach to monitoring maintenance perfor- ministrative staff units, and the head of the service.
mance in a highly complex hospital with vast and varied tech- A mixed maintenance strategy is in place due to the tech-
nology. Although the expression “evidence-based” is well nological complexity and the number and type of different
known in the medical literature, it may also be applied to pieces of equipment. Dedicated internal technicians take care
maintenance. Evidence-based maintenance (EBM) begins of scheduled and corrective maintenance of some classes of
with the analysis of evidence (i.e. failures) to monitor the equipment and are in charge of first-level maintenance in part-
maintenance effectiveness and plan any necessary changes ner agreements with manufacturers or distributors. Internal
to improve it. Maintenance reports in most hospitals describe maintenance is adopted for surgical lamps (LSC), ceiling-
only the failures, the repair procedures and any spare parts mounted units (PSO) and telemetry devices (UTC).
used. What these reports never provide is information about For critical or high-tech devices, maintenance is covered by
any measures needed to prevent that failure [21]. Knowledge full risk agreements with manufacturers or authorized service
of the history of a failure enables the monitoring and improve- centres. External maintenance is adopted for anaesthesia ma-
ment of the current maintenance strategy so that the most chines (ANS), central monitoring systems (CMO), electrocar-
appropriate approach can be found. Ultimately, when the ef- diographs (ECG), vital parameter monitors (MON), surgical
fectiveness, reliability and availability of medical equipment tables (TOP) and ventilators (VPO).
are improved through maintenance, the safety of staff and The maintenance of aspirators (ACH), defibrillators (DEF),
patients is improved. electro-surgery units (ELB) and oximeters (OOR) is entrusted
The objective of this study is to verify the feasibility of to a global service provider also in consideration of their
implementing an evidence-based method (i.e. based on the amount and the diversity of their manufacturers.
history of failures) for maintenance. In this way, through the The analysis in this paper concerns data from the equip-
study of current maintenance procedures, the steps required ment used in intensive care and surgery departments, includ-
for strategic maintenance policy changes can be applied. This ing vascular intervention. Table 1 shows the number of oper-
research paper is grounded on the EBM approach applied by ating rooms involved in the analysis and the number of inten-
Wang et al. [21–24]. sive care beds. Indeed, these departments, among the most
The first step of this process was to classify the mainte- critical in the hospital, are characterized by high technological
nance work orders (WOs) using a set of codes. The same small heterogeneity. The data refer to the period 2012–2016.
set of codes selected in [21] was used to standardize and sim- Figure 1 describes, in a block diagram, the process for
plify WO classification. Then, analysis of the SM and CM calculating the proposed set of KPIs, as detailed in this sec-
medical equipment records enabled the identification of un- tion. The process starts with research on data for technologi-
usually high code incidence and issues related to possible cal, organizational and financial KPIs that can be found in the
omissions. medical equipment database. Then, all the required data for
The second step was the design of a novel set of key per- CM, SM, inventory and information from maintenance con-
formance indicators (KPIs) useful for assessing the perfor- tracts are collected. Information concerning factors such as
mance of medical equipment maintenance. The most suitable costs, durations, number of devices and human resources is
indicators for the available data, information and context were meticulously selected also from server software, certified mail
selected among those available in the literature. and invoices as well as from direct interviews with techni-
Some prior papers related to the EBM approach and to the cians. The KPIs can be calculated and analysed graphically
use of KPIs for evaluating medical equipment maintenance and with the help of business intelligence software.
performance were presented by the authors at international
conferences [10–12, 17] or published in international journals 2.1 Data analysis of hospital equipment
[2].
The first step was to identify the classes of equipment as a
target for the focused research analysis.
It was decided to give relevance to the most numerous and
2 Materials and methods critical devices. The choice was restricted to classes of equip-
ment with more than 40 units belonging to the two aforemen-
This study began in December 2016 at Careggi University tioned departments. Table 1 shows the 13 selected classes with
Hospital, which has 1367 beds and 16,209 pieces of medical the device type in the first column and the quantity in the
equipment. Management of the medical equipment and its second. We analysed technical reports on CM and SM activ-
maintenance is entrusted to the Department of Information ities, including preventive maintenance, electrical safety tests
and Health Technology of the regional health service body and quality control. Table 1 also shows the number of CM and
ESTAR (Ente di Supporto Tecnico-Amministrativo SM WOs in the third and fourth columns. The rightmost three
Med Biol Eng Comput

Table 1 Analysed data

Departments Rooms Beds Wards

Operating rooms 40 9
Interventional 5 5
Intensive care 165 10
Device type TOT Units (U&oU) TOT CM WO TOT OR&IC OR&IC OR&IC SM WO
SM WO Units CM WO
Anaesthesia Machine 162 802 491 109 593 444
Aspirator 377 160 287 43 20 42
Ceiling mounted unit 319 284 522 214 165 386
Central monitoring 63 212 147 33 114 87
Defibrillator 410 1463 2036 128 438 709
Electrocardiograph 356 1384 947 57 155 148
Electrosurgical 205 287 408 148 181 342
Monitor 900 1294 3337 487 547 1794
Oximeter 613 557 1120 154 110 297
Surgical lamp 354 411 1222 225 239 987
Surgical table 93 520 382 70 349 211
Telemetry 104 99 142 27 59 51
Ventilator 203 796 831 155 611 748
Total analysed data 4159 8269 11,872 1850 3581 6246

columns show how many pieces of equipment are in use in 2.2 Failure classification
surgery departments and intensive care units, as well as the
related CM and SM work orders. The purpose of classifying the maintenance operations
A total of 14.06% of the data were excluded from the anal- was to analyse and monitor the types of performed oper-
ysis because the related reports lacked enough information for ations. The number of CM cases corresponds to the num-
a proper classification. ber of failures that occurred (except for false failures,

Fig. 1 The left side shows a block diagram of the process needed to scheduled maintenance analysis for a specific class of equipment is
calculate the proposed set of KPIs, starting from technological, shown (legend in the article body). The bottom right part shows how
organizational and financial data. On the upper right, an example of the KPIs can be implemented in a business intelligence dashboard
 Med Biol Eng Comput

NPF). The same codes used in [21] are used to identify Taking into account the criteria suggested by the above
each failure type summarized in Table 2. Each individual standard, as well as the analysed literature and our personal
CM and SM technical report was carefully analysed so knowledge and needs from the field, we designed a set of 20
that the many failures that occur each year could be KPIs, which are thoroughly described in Table 3 below. The
catalogued. In ambiguous cases, when there was a possi- table summarizes the information on the chosen indicators:
ble correspondence of two or more codes with the same their name, the type of indicator (financial, organizational or
failure, the most appropriate code was selected through technological), the mathematical definition and the rationale
careful analysis performed in cooperation with CE tech- behind it. Moreover, the table indicates which activities are
nicians and staff. pertinent to each indicator, between CM and SM. Internal
maintenance (IM) and/or external maintenance (EM) activities
are indicated for each indicator as well. The identified indica-
2.3 KPIs tors were calculated for each year from 2012 to 2016 for each
of the 13 chosen equipment classes to obtain the overall be-
The UNI EN 15341:2007 standard [19] describes a sys- haviour and evolution of each indicator over time.
tem for managing KPIs to measure maintenance perfor- The set of 20 indicators in Table 3 not only come from
mance as influenced by key maintenance factors and to the UNI EN 15341 standard but also are intended as a
assess and improve efficiency and effectiveness. The stan- novel research result of our study.
dard is applicable to many industrial and technical sectors. To further clarify the indicator concepts of downtime
The maintenance of medical devices must ensure equip- and uptime, the European standard EN 13306:2010 was
ment availability and reliability (linked to the safety of the used as a reference [18]. Downtime is the time interval
device). throughout which an item is not capable of performing its
The standard suggests that the KPIs be structured into function. Uptime is the time interval throughout which an
three groups to measure every aspect of the maintenance item is fully functional. The well-known mean time to
process. A thorough review of the literature led to the restoration (MTTR) and mean time between failures
selection of the three groups below to match the CE de- (MTBF) are the average times to restoration of function
partment’s data and requirements. These KPIs are as and the average time between consecutive failures,
follows: respectively.
With regard to the financial indicators, the acquisition
1. Financial, with the assessment of the cost-effectiveness of cost (used in KPI-15) was derived from the tables show-
the performance being the primary objective [1, 3, 6, 19] ing the purchase value estimates for each equipment class
2. Technological, with the assessment of the operational per- supplied by the CE department in 2014, increased by
formance of the equipment in terms of its reliability and 20%. Furthermore, a 2% increase or decrease was estimat-
availability (related to customer satisfaction) as its aim [1, ed for the years subsequent to and prior to 2014, respec-
13, 16, 17, 19, 25, 27] tively. The acquisition costs reached for each class were
3. Organizational, which is related to internal processes and multiplied by the annual number of each class. The deci-
staff productivity [1, 5, 10, 13, 14, 16, 17, 25, 27] sion to use purchase value estimates was justified, as

Table 2 Failure codes

Code Description CM/

SM

NPF No problem found Both

BATT Battery failure Both
ACC Accessory failure (including supplies) Both
NET Failure related to network CM
USE Failure induced by use (i.e. abuse, accident, environment conditions) CM
UPF Unpreventable failure caused by normal wear and tear CM
PPF Predictable and preventable failure CM
SIF Induced by service (i.e. caused by a technical intervention not properly completed or premature failures of a part just replaced) CM
EF Evident failure (i.e. evident to user but not reported) SM
PF Potential failure (i.e. in process of occurring) SM
HF Hidden failure (i.e. not detectable by the user unless special test or measurement equipment) SM
Table 3 Key performance indicators

Index KPI Definition Rationale Involved activity

type
Corrective Scheduled Internal External
Maintenance Maintenance Maintenance Maintenance
CM SM IM EM
Med Biol Eng Comput

KPI 1 T T down ð%Þ ¼ TRTnd 100 Operational efficiency, actual equipment X X

Downtime (%) (non-availability with: availability compared with
time) Tnd = non-availability time per year; requirements.
RT = Required Time per year.
KPI 2 T Td Operational efficiency, actual equipment X
T up ð%Þ ¼ RT 100
Uptime (%) (availability time) with: Td = RT − Tnd availability compared with
requirements.
KPI 3 T T Parameter of reliability, availability. X
MTTR ¼ N CMf
MTTR (mean time to Tf is the off-duty time for failure;
restoration) NCM is the total number of corrective actions.
KPI 4 T d Parameter of reliability, availability. X
MTBF ¼ NTCM
MTBF (mean time between Td is the availability time;
failures) NCM is the total number of corrective actions.
KPI 5 T Class Failure Ratio ¼ NN CMi Failure rate of each class of equipment X
CM
Class failure ratio (fails per NCMi is the number of corrective actions per year applied to the ith
class) equipment class;
NCM is the total number of corrective actions in the same year.
KPI 6 T GFR ¼ NNCM Fault occurrences related to the number X
dev
Global failure rate NCM is the total number of corrective actions per year; of devices
(defectiveness) Ndev is the number of devices in the inventory at the end of the year.
KPI 7 T N CM jage class Device obsolescence X
AFR ¼ N dev jage class
AFR: age failure rate
NCM is the total number of corrective actions per year;
Ndev is the device number.
Age classes: 0–2 years, 3–5 years,
 6–9  years, ≥10 years
KPI 8 O N negl Operational performance of maintenance X
Negligent Actionsð%Þ ¼ N CM 100Nnegl is the number of
“Negligent” actions (%) process
corrective actions per year, that have not been completed within
30 days (“negligent” actions);
NCM is the number ofcorrective
 actions per year.
KPI 9 O N 1day Operational performance of maintenance X
1day actionsð%Þ ¼ N CM 100
“1 day” actions process
N1day is the number of corrective actions per year,
that have been completed within 24 h;
NCM is the number of corrective
 actions  per year.
KPI 10 O SM with failureð%Þ ¼ N SM failure
100 Scheduled maintenance intervention X
SM with failure (%) N SM with fault occurred
NSM failure is the number of scheduled maintenance actions per year
with code ≠ NPF;
NSM is the number of scheduled maintenance
 actions per year.
KPI 11 O SM Coverage Rate ð%Þ ¼ N SM
100 Scheduled Maintenance conformity to X
SM coverage rate (scheduled N dev the requirements
maintenance) SM coverage rate
Table 3 (continued)

Index KPI Definition Rationale Involved activity

type
Corrective Scheduled Internal External
Maintenance Maintenance Maintenance Maintenance
CM SM IM EM

NSM is the number of scheduled actions per year;

Ndev is the number
 of devices
 available in that year.
KPI 12 O N °NPF No fault found during the corrective X
N °NPF ð%Þ ¼ N CM 100
No problem found (fake faults) maintenance work order
(%) NCM is the number of corrective actions per year.
No:device


KPI 13 O No:technicians
Maintenance workload X X
No. devices per technician
(internal)
KPI 14 O Working hours spent on corrective maintenance vs working hours Maintenance-workload comparison X X
Time cost of the workforce spent on scheduled maintenance between corrective and scheduled
 maintenance
KPI 15 F Global Maintenance Cost Maintenance service: financial X X X X
COSRð%Þ ¼ Acquisition Cost Þ100
COSR (cost of service performance (cost-effectiveness).
ratio = global maintenance
to acquisition cost) (%)
KPI 16 F External Maintenance Cost Impact of external maintenance on the X
Total Maintenance Cost Þ100
External maintenance Cost where external maintenance cost = scheduled and corrective external total cost of the maintenance service
(% with respect to total maintenance costs
maintenance cost)
KPI 17 F Internal Maintenance Cost Impact of internal maintenance on the X
Total Maintenance Cost Þ100
Internal maintenance cost where internal maintenance cost = scheduled and corrective internal total cost of the maintenance service
(% with respect to total maintenance costs
maintenance cost)
KPI 18 F Corrective Maintenance Cost Maintenance type: impact of corrective X
Total Maintenance Cost Þ100
Corrective maintenance cost where corrective maintenance cost = internal CM cost + external CM maintenance on the total cost of the
(CM) (% with respect to total cost maintenance service.
maintenance cost)
KPI 19 F Scheduled Maintenance Cost Maintenance type: impact of scheduled X
Total Maintenance Cost Þ100where scheduled maintenance
Scheduled maintenance cost cost = internal SM cost + external SM cost maintenance on the total cost of the
(SM) (% with respect to total maintenance service.
maintenance cost)
KPI 20 F Cost of Spare Parts Maintenance: Spare Parts and X X X X
Total Maintenance CostÞ100
Cost of spare parts consumables.
(+ consumables)
(% with respect to total
maintenance cost)
Med Biol Eng Comput
Med Biol Eng Comput

Fig. 2 CM and SM failure distributions related to surgical tables, telemetry equipment, electro-surgery units and defibrillators in different years of the
considered period

indicated in the literature [8], by the fact that the purchase Therefore, the total maintenance cost was calculated using
cost of each individual device represents only the initial these cost items:
portion of the total cost of ownership of that device. The
total cost of ownership comprises several cost items, such – Contract costs
as contracts, spare parts, accessories, consumables and – Costs for spare parts and consumables
instruments used to perform test measurements. – Costs for batteries
To calculate total maintenance costs, only the cost – Costs for internal maintenance personnel
items clearly linked to each device were considered. – Extra-contractual costs (all costs not elsewhere covered)
 Med Biol Eng Comput

3 Results lower rate of scheduled maintenance coverage than the rate

provided by the maintenance plan before 2016.
This section displays the analysis and the graphs derived from Another interesting fact emerged by comparing the CM
the methods described in the previous section. Given the large and SM fault patterns for defibrillators (Fig. 2), the latter in-
amount of data, only the graphs related to some of the devices cluding preventive maintenance, electrical safety audits and
analysed are included. The histograms show a stable pattern quality control. The classes with potential maintenance omis-
when more than 50 sets of data are used, which is in line with sions are PF and EF in SM and PPF in CM. The latter category
the literature [21]. By analysing all the equipment classes used is responsible for approximately 3% of all failures. Through
in the hospital, a characteristic performance shape may be the analysis of the technical reports, this circumstance may be
obtained for this specific hospital, which can then be used to ascribed to poor first-level maintenance by health personnel.
make comparisons/benchmarks with other hospitals. This result is in line with the peculiarity of Careggi University
Hospital, which is an extremely large healthcare facility with a
very high rate of personnel turnover. This could affect staff
3.1 Distribution of classified failures accountability in asset management (the conflict between uni-
versity and hospital property) reflected in the consistency of
The histograms in Fig. 2 show the distribution of failure codes the USE failure class incidents that include accidental failures
obtained from CM and SM WOs, which were related to sur- and failures due to the misuse of equipment. These data con-
gical tables, telemetry equipment, electro-surgery units and firm the importance of staff training. Since there is no user-
defibrillators. The graphs obtained through the analysis of training programme beyond initial training during device test-
CM WOs are in the first column. The graphs from SM WOs ing, there is a high incidence of failures due to improper de-
are in the second column. Details are sometimes shown for a vice management.
better data comprehension. From the corrective maintenance pattern for electro-sur-
The first five histogram bars for each type of failure repre- gery units (Fig. 2), it is evident that the most affected category
sent the five investigated years (2012 to 2016). The rightmost in this case is UPF. This type includes a broader range of
one is the average value with error bars of ± 1 standard devi- failures, normally attributed to wear. The scheduled mainte-
ation (SD). The height of the bars represents the percentage of nance pattern shows a prevalence of procedures with positive
failures found in CM or SM WOs. outcomes (NPF). Nonetheless, there is a 1% potential failure
The CM values were corrected using the equipment type rate, which might be related to issues that can be resolved by
failure rate (ETFR), which is the percentage of units within a increasing the frequency of scheduled maintenance and by
specific equipment type that failed each year [21]. This cor- paying more attention to checking the correct function of com-
rection is necessary when a combination of CM and SM fail- ponents that are more prone to failure (pedal, plates and
ure code distributions is required to provide a more complete handpieces).
view of the equipment fault history. By analysing the CM procedures on surgical tables from
In fact, the CM WOs are only related to failed units and did 2012 to 2016 (Fig. 2), the most significant category of failures
not consider units that had not failed. Instead, the SM WOs is ACC (failures of accessories). Some of the indirect actions
refer to all units. the CE department could implement to reduce this type of
From the top left chart in Fig. 2 (CM distribution for defi- failure belong to the procurement stage. Giving importance
brillators), some interesting information can be gleaned about to the reliability of accessories and spare parts as well as
the effectiveness of the most common types of maintenance analysing the brands in use with a higher failure rate in relation
procedures used in the category of defibrillators. It appears to the total number of units in the inventory could be effective
that the most significant failure category involves batteries in reducing these failures. The scheduled maintenance chart
(BATT). Specifically, in 2015, 60.40% of CM WOs were for clearly shows that, every year, scheduled maintenance proce-
battery failures, 8.91% were related to wear and 5.94% were dures on surgical tables take place with no negative results
unpreventable failures. In 2016, 93.33% of the SM WOs were reported. The remaining 2% of the maintenance WOs were
NPF, 3.81% were for BATT and 2.86% were preventable fail- coded as evident failures (EF). To reduce this type of failure,
ures (PF). personnel should be trained to immediately report failures and
From a review of the literature [23], a comparison of main- problems that are evident and can be identified with no special
tenance strategies in different hospitals shows consistent dif- tools or measurements.
ferences. As expected, BATT failures were lower in cases The telemetry corrective maintenance histogram for fail-
where scheduled maintenance was performed more frequent- ures from 2012 to 2016 shows a failure history that includes
ly. The paper confirms that a higher frequency of scheduled three types of problems: UPF, USE and NPF. The peak of the
maintenance reduces certain types of failures (BATT). UPF category for 2016 is in line with the list published by
Therefore, the pattern from our case could be explained by a ECRI for the “Top 10 Health Technology Hazards” for 2016
Med Biol Eng Comput

[9]. This report states that telemetry failures and the resulting an average of 69.48% of the total cost. This is an example of
lack of monitoring of a patient’s vital signs are in 4th place how evidence-based maintenance works. Because of the ex-
among the hazards to patients from medical technology fail- perience of internal technicians, starting from the evidence, it
ures. This report mainly discusses the improper use of medical was possible to adjust the maintenance policy, leading to eco-
devices. Moreover, it highlights the importance of taking ac- nomic improvement.
tions focused on reducing USE because this is the other cate- A comparison of the cost patterns for KPI 18 (CM cost) and
gory of failures that affects these devices. For example, among KPI 19 (SM cost) on a single class of equipment (telemetry)
the indirect actions the CE department could implement, it is showed that preventive maintenance accounted for an average
worth reiterating that better staff training can be effective in of 5.39% of the cost compared with corrective maintenance,
appropriate telemetry management. which accounted for 94.61%. This difference can be explained
From a review of the technical reports on telemetry, the by considering the 0.56 average coverage rate of scheduled
high level of stress on these devices emerges, as they are maintenance for this class. Therefore, to improve this situa-
constantly connected to patients being monitored. The pur- tion, the maintenance policy should be changed to guarantee
chase of more robust and reliable equipment that is able to that preventive maintenance will be performed on each device
better handle high technological stress levels could be a solu- at least annually. This improvement in the maintenance sched-
tion. The scheduled maintenance procedures on these devices ule would probably reduce corrective maintenance costs.
have a 100% success rate with all positive outcomes (NPF). Among the most unusual cost patterns for spare parts (KPI
20) were those for surgical tables and pulse oximeters, whose
3.2 Assessment using KPIs average values were 34.05% and 34.82%, respectively.
Indeed, these classes of equipment have accessories and spare
This section presents some considerations on the values of the parts that wear so rapidly that maintenance is closely linked to
set of indicators identified in Section 2.3. their hours of use; they often fail and are replaced. On average,
All the calculated KPIs for each equipment class, as well as for the other classes, there is a percentage incidence of 30%
the distribution of the equipment in classes of age, are shown for SM and a percentage incidence of 60% for CM. The re-
in Tables 4, 5, 6a and b and 7 below. maining portion is attributable to the costs of the spare parts,
Concerning the economic KPIs, the COSR (cost of service affected with higher or lower relevance depending on the type
ratio) results (i.e. KPI 15) were compared with the values of of contract: full risk contracts, for example, include all the
the economic indicators proposed by the Procurement Unit of spare parts in the annual fee. For equipment with this type of
the Italian Public Administration CONSIP (i.e. Public contract, including monitors and monitoring stations, the cost
Information Services Licensee). The value estimates were in of spare parts, calculated as a separate item, is very low.
line with the CONSIP values [7]. In this comparison, the Comparison between downtime with no negligent mainte-
electro-surgery unit class deserves further investigation be- nance service and downtime (KPI 1) due to negligent actions
cause it has an average COSR value (1.45%) lower than that (KPI 8)—i.e. service interventions lasting more than
of CONSIP. Until 2014, there was a no-cost maintenance ser- 30 days—showed that, for all classes, negligent maintenance
vice policy in place for many of these devices (by contract, the service led to a considerable increase in downtime (2%, on
cost was absorbed by the purchase of consumables); this fact average), with notable patterns for surgical tables, anaesthesia
strongly influenced this value. In addition, the range of the and telemetry.
electro-surgery technology was highly variable, requiring ex- For the first and the second equipment, downtime was af-
tremely specialized equipment, with high initial purchase fected significantly by the time required for spare parts to be
costs and very high consumable costs. CONSIP estimates delivered. Indeed, for these categories, failures of accessories
put electro-surgery maintenance incidence at a medium-high (ACC) were significant. Any equipment downtime directly
level of 8%, so it seems clear that a policy with a service affects its availability or uptime (KPI 2). The uptime pattern
formula would be preferable to purchase. for surgical tables with and without negligent maintenance
The results related to KPI 16 (external maintenance cost) service is shown in Fig. 3. Clearly, negligent maintenance
and KPI 17 (internal maintenance cost) for the class of elec- service significantly affected uptime.
trocardiographs, which are characterized by both internal and For the anaesthesia machine, it is also necessary to consider
external maintenance policies, should be highlighted due to that the SM has a long duration for each intervention and,
the impact of this mixed policy on total maintenance costs. considering that this class has an average SM coverage rate
From 2012 to 2014, external maintenance costs accounted for of 1.09, the time dedicated to preventive maintenance affects
an average of 93.03% of the total cost. Beginning in 2015 the availability of the equipment (13.6%, on average) and,
through 2016, the maintenance policy changed and internal consequently, the uptime. Instead, the problem with telemetry
technicians provided maintenance. Therefore, between 2015 could be linked to a higher incidence of negligent maintenance
and 2016, the impact of external maintenance costs dropped to service. This leads to a hope for an organizational correction
 Med Biol Eng Comput

> 10 years

± 0.3

± 0.1

± 0.1
± 0.0
± 0.2

± 0.0

± 0.1
± 0.2
± 0.5

± 0.2

± 0.5
± 0.1

± 0.1
of the maintenance policy. Specifically, the implementation of
an accurate monitoring system for service calls protracted over

1.3

0.5

0.3
0.2
0.1

0.0

0.1
1.1
0.8

0.2

1.2
1.1

0.5
time, such as a dashboard, could reduce negligent mainte-
nance service that prolongs downtime and delays the avail-

5–10 years

3.2 ± 1.0

0.3 ± 0.1

0.1 ± 0.1

0.5 ± 0.1
1.3 ± 0.4

0.3 ± 0.4

0.2 ± 0.1
0.6 ± 0.2
1.9 ± 0.5

0.0 ± 0.1

1.3 ± 0.3
1.8 ± 0.5

0.5 ± 0.3
ability of the device.
The average annual uptime value, directly related to the
downtime figure, for all equipment is better than 94%,
2–5 years which is the CONSIP figure that should be guaranteed

1.9 ± 1.3

0.5 ± 0.4

0.7 ± 0.6
0.3 ± 0.2
0.7 ± 0.2

0.6 ± 0.6
1.9 ± 0.7
0.4 ± 0.1

0.4 ± 0.4

0.3 ± 0.1
0.2 ± 0.3
1.4 ± 0.8

0.9 ± 0.9
each year.
AFR: age failure rate

It is no surprise then that the class with the highest average

Mean 5Y ± SD

MTTR (KPI 3) values (approximately 4.5 days) is “surgical

0–2 years
Age class

tables,” which, as mentioned above, are affected by negligent

1.0 ± 0.4

0.0 ± 0.1

0.1 ± 0.1
0.1 ± 0.1
0.5 ± 0.5

0.0 ± 0.0

0.1 ± 0.1
1.2 ± 0.7
0.2 ± 0.1

0.2 ± 0.2

0.2 ± 0.1
0.6 ± 0.4

0.9 ± 0.5
KPI 7

maintenance service. The time required to restore the correct

function of the devices in this category is strictly linked to the
time required for spare parts shipping. The surgical tables’
Mean 5Y ± SD
Global failure

MTTR pattern in Fig. 4 shows a peak in 2014 for the MTTR

0.36

0.13

0.04
0.05
0.26

0.12
0.18
0.07

0.25

0.06
0.06
0.12

0.16

with negligent maintenance service. Instead, the MTTR with-

1.35 ±

0.30 ±

0.29 ±
0.16 ±
0.87 ±

0.64 ±
1.45 ±
0.41 ±

0.96 ±

0.20 ±
0.12 ±
1.07 ±

0.53 ±
KPI 6

out negligent maintenance service is at the minimum uptime

rate

for that year.

The MTBF (KPI 4) should be as high as possible. An
Mean 5Y ± SD

2.52

2.54

1.21
1.59
0.89

0.48
1.92
1.49

2.30

1.18
0.42
1.83

1.54

acceptable figure is one failure every 6 months (approximately

Class failure

4500 h). Surgical tables and anaesthesia have the worst (i.e.
±

±
±
±

±
±
±

±
±

±
KPI 5

9.59

6.63

15.33
4.55
3.13

1.67

3.13
16.59
5.14

0.58

12.19
17.10

4.37

lowest) values. These values agree with the global failure rate
ratio

(KPI 6) figures. A low global failure rate is generally associ-

ated with equipment with low technological complexity. In
(without negligent

fact, operating tables and anaesthesia have the lowest MTBF

work orders)

values (1 failure every 2 months), while monitors, pulse

(months)

oximeters and aspirators have the highest. These values are

MTBF
KPI 4

1.91

8.81

15.43

38.04
14.10

15.26
1.74

12.63

34.56
6.22

20.38
9.04

11.74

in agreement with the global failure rate, which is the lowest

for these three categories of equipment, while it is greater than
1 (more than one fault per year) for both operating tables and
(without negligent

anaesthesia. This result indicates that operating tables and an-

work orders)

aesthesia have a greater defectiveness than, for example, pul-

monary ventilators (average failure rate of 1.07). The low
MTTR

(days)
KPI 3

KPI6 for aspirators is in line with the low technological com-

4.70

4.24

2.41

3.73
2.68

3.05
3.02
2.59

1.71

1.83
2.30
3.09

2.85

plexity of these devices and their low impact on maintenance

costs. For pulse oximeters and monitors, medium-high classes
(without negligent

0.69%

0.44%

0.12%

0.05%
0.36%

0.27%
0.61%
0.30%

0.36%

0.08%
0.16%
0.40%

0.33%

on the cost of maintenance, a low KPI6 indicates fewer faults

Mean 5Y ± SD
work orders)

in the equipment itself.

Uptime (%)

±
±

±
±
±

±
±
±

The class failure ratio index (KPI 5) indicates that pulmo-

97.26%

99.36%

99.79%

99.69%
99.38%

99.48%
97.86%
99.51%

99.53%

99.87%
99.87%
99.02%

99.46%
KPI 2

nary ventilators and anaesthesia machines are the classes that

have most affected the failure rates in operating rooms and
intensive care units over the last 5 years. Although the lung
(without negligent

ventilators are the class that most greatly affects the total num-
0.69%

0.44%

0.12%

0.05%
0.36%

0.27%
0.61%
0.30%

0.36%

0.08%
0.16%
0.33%
Technical KPIs

Mean 5Y ± SD
Downtime (%)

0.4%
work orders)

ber of faults, compared with their number, they have a lower

defectiveness with respect to the anaesthesia machine. In ad-
±

±
±

±
±
±

±
±
±
±
2.74%

0.64%

0.21%

0.31%
0.62%

0.52%
2.14%
0.49%

0.47%

0.13%
0.98%
0.13%
0.54%
KPI 1
Technical KPIs

dition, ventilators and anaesthesia are the classes that most

affect the cost of maintenance, having average COSRs of
8.48% and 5.30%, respectively. Therefore, when planning
Equipment class

maintenance strategies, a balance must be found between

two key aspects. On the one hand, costs need to be contained;
Table 4

on the other hand, the criticality of the equipment must be

MON
CMO

ACH

OOR
UTC
ANS

VPO

ECG
ELB

DEF
TOP

LSC

PSO

considered. For example, even if anaesthesia machines have

Med Biol Eng Comput

Table 5 Distribution of the equipment in age classes

Age classes (mean no. of units for each age class)

Mean 5Y ± SD

Equipment class 0–2 years 2–5 years 5–10 years > 10 years

CMO 28.49% ± 16.48% 31.28% ± 20.84% 31.33% ± 13.90% 8.90% ± 4.22%

TOP 32.77% ± 12.76% 13.64% ± 14.79% 11.64% ± 2.86% 41.94% ± 5.55%
UTC 0.00% ± 0.00% 38.00% ± 35.64% 38.74% ± 45.63% 23.26% ± 13.10%
ANS 26.43% ± 14.49% 25.33% ± 6.98% 19.32% ± 3.48% 28.92% ±6.77%
ELB 24.74% ± 3.72% 24.51% ± 4.55% 19.12% ± 5.58% 31.63% ± 4.11%
VPO 10.78% ± 5.63% 23.37% ± 18.27% 23.93% ± 11.09% 41.92% ± 3.14%
ACH 9.95% ± 6.01% 5.13% ± 5.40% 15.17% ± 6.93% 69.75% ± 6.53%
ECG 9.96% ± 8.23% 10.05% ± 7.45% 21.03% ± 3.26% 58.96% ± 6.66%
LSC 33.97% ± 21.25% 20.38% ± 20.73% 12.75% ± 5.17% 32.90% ± 3.21%
DEF 10.63% ± 4.91% 13.72% ± 8.91% 33.67% ± 4.65% 41.98% ± 6.04%
PSO 42.34% ± 29.29% 25.64% ± 28.00% 10.01% ± 2.47% 22.00% ± 0.90%
OOR 23.38% ± 9.96% 28.21% ± 12.41% 27.20% ± 9.96% 21.21% ± 5.59%
MON 33.02% ± 20.56% 25.81% ± 17.47% 22.01% ± 5.57% 19.16% ± 2.93%

a high impact on maintenance costs, these pieces of equipment age do not always show a higher failure rate than the classes
cannot receive less maintenance because they have a high with newer devices.
failure rate and are vital. The age indicator pattern suggests that the variability in age
Contrary to expectations, an analysis of the age failure rate in terms of failure rate is probably much less than the variabil-
(KPI 7) shows no correlation between failure rate and obso- ity of other factors, such as whether operators manage and use
lescence. This could be due to the presence of no problem the devices properly or not. Therefore, age does not seem to be
found (KPI 12, i.e. “fake faults”) that do not represent real a significant parameter in current maintenance policies.
failures of the equipment and represent a good 19% of the Negligent maintenance service (KPI 8) (corrective mainte-
total corrective interventions (684 out of 3581). This category nance calls resolved in more than 30 days) are to be applied
could introduce a distortive component (bias) that prevents a together with uptime (KPI 2), downtime (KPI 1) and MTTR
clear interpretation of the indicator performance. The calcula- (KPI 3). Compared with the total number of failures, the te-
tion of this indicator was, therefore, repeated by removing the lemetry equipment and surgical table classes have the highest
percentage attributable to NPF for some classes, but in gener- number of negligent maintenance service calls. Instead, the
al, a correlation between failure rate and age of the device did number of 1-day service calls (KPI 9) (corrective maintenance
not appear evident. However, to calculate KPI 7, the equip- service requests resolved within 24 h) was found to be lower
ment was divided into age classes, and this subdivision pro- for telemetry equipment, which were more affected by negli-
vided useful information on the age composition of the oper- gent maintenance service. However, 64.49% of defibrillators
ating room and intensive care equipment (see Table 7). The were serviced within 24 h. This equipment also has KPI 1 and
data could then be compared with the average age from the KPI 3.
literature. For example, among the analysed classes of equip- No problem found (KPI 12) was more significant for aspi-
ment, it was found that defibrillators, ECGs, aspirators, venti- rators and anaesthesia machines. However, if compared with
lators and operating tables comprise more than 40% of devices the involved workload (approximately 5 days a year), it does
that are older than 10 years. By consulting the data from the not greatly affect the wasted time. In this case, investing in
Biomedical Engineering Advisory Group (BEAG) [4] and the training to instruct staff to open corrective interventions in a
American Hospital Association (AHA) [26], it can be noted better way would not be economically advantageous because
that the average age of operating tables and vacuums is equal the incidence of these “fake faults” does not justify the
to 15 years, that of ventilators and ECG ranges between 7 and investment.
10 years, and that of defibrillators is 5–7 years. Therefore, in The SM coverage rate (KPI 11) is higher than 1 SM inter-
our data, defibrillators have an average higher age with respect vention per year for surgical lamps, ventilators and anaesthe-
to the values reported by BEAG and AHA. From the analysis sia machines. For defibrillators, only in 2016 is the planned
of the failure pattern, however, the classes of equipment in target of 2 preventive maintenance interventions per year
which there is a high percentage of devices over 10 years of reached because of the transition from internal service to
 Med Biol Eng Comput

Table 6 Organizational KPIs

Organizational KPI
a
Equipment class KPI 8 KPI 9 KPI 10 KPI 11 KPI 12
“Negligent” actions (%) “1 day” actions SM with failure (%) SM coverage rate No problem found (%)
Mean 5Y ± SD Mean 5Y ± SD Mean 5Y ± SD Mean 5Y ± SD Mean 5Y ± SD
CMO 7.51% ± 10.52% 42.25 ± 17.41 1.00% ± 2.24% 0.67 ± 0.07 16.97% ± 5.73%
TOP 10.64% ± 6.63% 22.35 ± 5.97 1.83% ± 3.04% 0.84 ± 0.30 14.05% ± 4.73%
UTC 55.63% ± 12.48% 9.14 ± 11.72 0.00% ± 0.00% 0.56 ± 0.31 4.40% ± 4.02%
ANS 4.39% ± 4.84% 45.05 ± 9.19 1.82% ± 1.31% 1.09 ± 0.15 26.12% ± 4.93%
ELB 5.25% ± 4.21% 45.83 ± 24.69 1.00% ± 2.24% 0.77 ± 0.15 26.51% ± 9.97%
VPO 7.35% ± 3.74% 49.65 ± 11.27 0.87% ± 0.84% 1.31 ± 0.21 20.18% ± 2.92%
ACH 0.00% ± 0.00% 58.83 ± 25.71 0.00% ± 0.00% 0.30 ± 0.26 38.00% ± 41.47%
ECG 2.78% ± 2.89% 47.23 ± 19.52 0.83% ± 1.86% 0.51 ± 0.25 19.85% ± 6.42%
LSC 4.21% ± 1.39% 51.61 ± 15.00 1.54% ± 1.84% 1.18 ± 0.51 13.03% ± 9.50%
DEF 3.81% ± 2.93% 64.49 ± 21.66 4.17% ± 3.85% 1.56 ± 0.64 12.73% ± 2.49%
PSO 5.84% ± 2.32% 39.44 ± 7.99 1.19% ± 2.00% 0.38 ± 0.38 21.27% ± 9.35%
OOR 2.43% ± 3.33% 46.87 ± 27.29 4.32% ± 5.44% 0.54 ± 0.16 5.67% ± 1.92%
MON 12.91% ± 4.29% 29.08 ± 7.30 1.79% ± 1.42% 0.94 ± 0.28 22.53% ± 7.41%
b
Type of maintenance KPI 13 KPI 14
No. of devices per technician Time cost of the workforce
– (h)
Internal maintenance Service Internal maintenance Service
CM 221 137 5412.50 906.35
SM 221 82 2247.70 348.90

external global service, due to a low internal service coverage on what has been discussed so far, to improve or correct the
rate in previous years. Other classes of equipment have a value maintenance policy, it may be assumed that one ought to start
of KPI 11 less than 1 mainly because devices that cannot be with the evidence (i.e. the failure data). Planning a certain
found or that are used continuously are unavailable for main- maintenance strategy is not enough. Instead, it is necessary
tenance activities. Ceiling-mounted units from 2012 to 2015 to continuously monitor the behaviour of the key parameters
have KPI 11 values less than 0.4. This low value is due to an with the greatest impact on equipment availability (uptime,
SM frequency of 2 years, as specified in technical manuals. In negligent maintenance service calls, MTTR). This will not
Table 8, the yearly SM coverage rates for each equipment only optimize available resources but also improve the effec-
class are shown from 2012 to 2016. tiveness of the maintenance service and ultimately improve
By comparing internal maintenance and external service in patient and operator safety.
terms of number of devices per technician (KPI 13) and time
cost of the workforce (KPI 14) in CM and SM, it can be
concluded that, in relation to the considered equipment, the 3.3 Combined use of the maintenance service codes
internal technicians manage a higher number of devices than and KPI
the service at the expense of more hours spent in maintenance.
Despite a higher workload, the internal technicians mainly The combined use of the two approaches discussed thus far
manage classes of equipment such as the operating room cab- provides a tool for broad spectrum monitoring of the mainte-
inets, telemetries and scialytic lamps. They are highly special- nance process. Indeed, the KPI values can be better investi-
ized in these types of equipment; hence, they are able to opti- gated and understood by making use of the types of coded
mize their timing and manage a greater workload. maintenance calls. Similarly, if an unusually high failure type
The SM with failures (KPI 10) shows that, on average, SM is detected, the effects of this problem on the performance of
does not lead to the detection of failures (in fact, the mean the entire maintenance process may be observed. In this way,
value for the 5 years is less than 12%). Therefore, SM does targeted corrective actions can be taken to improve mainte-
not always succeed in intercepting failures or problems. Based nance service.
Med Biol Eng Comput

Table 7 Financial KPIs

Equipment Financial KPI

class
KPI 15 KPI 16 KPI 17 KPI 18 KPI 19 KPI 20
COSR (cost of service External Internal maintenance Corrective maintenance Scheduled maintenance Cost of spare parts
ratio = global maintenance cost cost (% with respect to cost (CM) (% with respect cost (SM) (% with respect (+ consumables)
maintenance to (% with respect to total maintenance to total maintenance to total maintenance (% with respect to
acquisition total maintenance cost) cost) cost) total maintenance
cost) (%) cost) cost)
Mean 5Y ± SD Mean 5Y ± SD Mean 5Y ± SD Mean 5Y ± SD Mean 5Y ± SD Mean 5Y ± SD

CMO 2.31% ± 0.37% 98.71% ± 0.41% 1.19% ± 0.51% 65.02% ± 0.48% 34.88% ± 0.55% 0.10% ± 0.13%
TOP 2.32% ± 0.20% 63.52% ± 14.44% 2.55% ± 1.04% 94.61% ± 3.06% 5.39% ± 3.06% 33.93% ± 13.72%
UTC 2.99% ± 1.56% 65.61 ± 41.17% 34.39% ± 41.17% 59.15% ± 1.91% 31.85% ± 1.03% 9.00% ± 2.94%
ANS 5.30% ± 0.84% 91.00% ± 2.94% –* 73.91% ± 16.13% 17.44% ± 9.93% 6.79% ± 8.04%
ELB 1.45% ± 0.81% 51.90% ± 0.82% 41.31% ± 15.18% 58.83% ± 2.87% 33.02% ± 2.61% 8.87% ± 4.53%
VPO 8.48% ± 0.35% 91.02% ± 4.62% 0.14% ± 0.19% 63.33% ± 26.73% 31.96% ± 26.41% 4.16% ± 7.55%
ACH 2.30% ± 2.04% 95.30% ± 7.55% –* 56.73% ± 9.92% 29.42% ± 5.11% 13.77% ± 14.70%
ECG 2.56% ± 0.40% 85.03% ± 15.91% 1.20% ± 0.97% 42.13% ± 16.86% 28.14% ± 12.46% 29.73% ± 13.94%
LSC 2.44% ± 0.87% 4.33% ± 4.53% 65.94% ± 14.12% 27.55% ± 6.39% 10.30% ± 3.18% 62.46% ± 8.25%
DEF 1.60% ± 0.71% 37.85% ± 8.25% –* 76.96% ± 8.13% 18.05% ± 9.16% 4.96% ± 2.58%
PSO 0.68% ± 0.40% 52.14% ± 8.53% 42.87% ± 8.54% 46.76% ± 26.46% 16.09% ± 7.60% 34.82% ± 27.02%
OOR 4.33% ± 2.49% 20.56% ± 33.76% 43.29% ± 22.53% 64.78% ± 1.32% 29.42% ± 4.37% 1.63% ± 0.96%
MON 3.61% ± 0.81% 96.36% ± 1.68% 2.00% ± 0.83% 65.02% ± 0.48% 34.88% ± 0.55% 0.10% ± 0.13%

*Fully external maintenance

For example, if surgical tables have above-average down- prolonged period out of service, but also maintenance costs,
time and MTTR, this can be investigated in greater depth by which represent the greater part of the cost items.
analysing the evidence, or rather, the types of failures. The Another analysis procedure could be performed beginning
spare parts cost indicator also shows a high impact on total with the classified service calls. Through a review of the cat-
maintenance costs for the surgical tables. From an analysis of egories with unusually high failure, the reasons for this singu-
the coded maintenance calls, one can find, that for this cate- lar behaviour may be investigated in more detail using KPIs.
gory of equipment, accessory failures have a greater inci- For example, electro-surgery units showed a peak of compo-
dence. The actions taken to improve the process can include nent failures in 2016. An analysis of the coverage rate indica-
monitoring the time needed for spare parts shipping and taking tor for scheduled maintenance calls showed that, in 2015 and
a survey to identify the most failure-prone accessories (models 2016, SM coverage was not optimal. In addition, the analysis
and brands) for consideration during procurement. of the coded corrective maintenance calls in 2016 showed that
For example, on average, 34.39% of costs were for internal 100% of the calls were classified as NPF. This seems to sug-
telemetry maintenance. More than half of all maintenance costs gest that specific checks on the components that tend to fail
were for repairs that required sending the device back to the most during scheduled maintenance calls should be included.
company for service. For this equipment class, the influence of Another example concerns the defibrillators, where an analysis
negligent maintenance service (maintenance calls that last more of the fault types revealed a prevalence of battery failures (BATT).
than 30 days) affects not only equipment availability, which is a An analysis of the scheduled maintenance indicator showed poor

Fig. 3 Surgery table: uptime

pattern with (red-dashed line) and
without (purple line) negligent
maintenance service
 Med Biol Eng Comput

Fig. 4 Surgery table: MTTR

pattern with (red-dashed line) and
without (purple line) negligent
maintenance service

coverage for defibrillators, which improved in 2016, the year the adoption or modification of a certain strategy. Starting from
equipment was transferred to a global service policy. the adoption of the evidence-based approach, the hospital
However, in 2016, the global failure rate for the defibrilla- could monitor the changes over time and compare them with
tor class was the highest of the 5 years analysed. This sug- the results of the previous maintenance policy. Transversal
gested that, despite adequate scheduled maintenance cover- comparisons examine other hospital circumstances.
age, it might be useful to include specific battery status checks The classification of maintenance service calls has
during the scheduled inspections. Given that 67% of the total highlighted that the main problem to be overcome is the need
maintenance costs for this class of equipment were represent- for descriptions of the maintenance work done that are as
ed by battery costs, appropriate battery management could accurate as possible. To a certain extent, by using failure
also reduce maintenance costs. codes, it may be possible to be freed of the need for detailed
descriptions of the maintenance service calls to track the spe-
cific type of failure or problem found.
By introducing the classification and analysis of maintenance
4 Discussion service calls as a part of the daily duties of maintenance techni-
cians, a valuable new tool that characterizes the equipment classes
By analysing the classes of equipment with the fault codes
in terms of problems and failures found can be implemented. The
proposed in this study, it is possible to understand the types
implementation of this approach could very well lead to greater
of problems most frequently encountered, which could be
optimization of the use of human and technological resources.
useful for longitudinal and transversal comparisons.
It should be noted that the adoption of the identified per-
Longitudinal comparisons involve the analysis of a certain
formance indicators provides a dual function in terms of the
maintenance policy at a given hospital before and after the
assessment and control of maintenance process performance
Table 8 SM coverage rate
and data communication and sharing. The data and informa-
tion conveyed by a hospital dashboard can provide summa-
Equipment class KPI 11–SM coverage rate rized yet complete output documents, which can represent
objective support for upper management decisions.
2012 2013 2014 2015 2016
The set of performance indicators defined concerned tech-
CMO 0.71 0.59 0.67 0.63 0.77 nological, organizational and financial aspects. The problems
TOP 1.10 0.43 0.69 0.87 1.14 encountered with the technology indicators were related to
UTC 0.50 0.95 0.40 0.16 0.79
ANS 1.01 0.99 0.95 1.30 1.18
scheduling. In fact, the data had to be updated in real time to
ELB 0.87 0.80 0.93 0.57 0.68 enable the assessment of the actual availability of equipment
VPO 1.14 1.35 1.30 1.11 1.63 and troubleshooting times.
ACH 0.05 0.23 0.38 0.14 0.71
The problems with the organizational and financial indicators
ECG 0.53 0.61 0.58 0.40 0.45
LSC 0.48 1.05 1.32 1.17 1.89 were similar. Essentially, these consist of the lack of a single
DEF 0.97 1.56 1.48 1.17 2.63 source for the data. To implement a set of indicators displayed
PSO 0.30 0.25 0.32 0.37 0.68 in a hospital dashboard, the indicators must be updated from a
OOR 0.44 0.73 0.32 0.56 0.65
MON 0.71 0.61 0.99 1.13 1.26 single data source. However, the fact that a complete financial
analysis requires the consultation of several sources, which
Med Biol Eng Comput

belong to structurally different processes and therefore cannot be creativecommons.org/licenses/by/4.0/), which permits unrestricted use,
distribution, and reproduction in any medium, provided you give appro-
merged into a single database, should also be considered.
priate credit to the original author(s) and the source, provide a link to the
Creative Commons license, and indicate if changes were made.

5 Conclusion
References
Maintenance is a crucial aspect of the activities in a hospital’s CE
1. AAMI (2016) Benchmarking solutions - HTM. In: products & publi-
department because it involves significant human and financial cations. AAMI. Available: http://www.aami.org/productspublications/
resources. The assessment of the effectiveness of any maintenance index.aspx?navItemNumber=505 Accessed 27 May 2019
programme is fundamental to the optimization of the use of avail- 2. Badnjevic A, Gurbeta L, Jimenez ER, Iadanza E (2017) Testing of
able resources in CE departments. In this respect, an approach for mechanical ventilators and infant incubators in healthcare institu-
tions. Technol Health Care 25(2):237–250. https://doi.org/10.3233/
monitoring maintenance from multiple points of view can be im- THC-161269
plemented by combining the classification of maintenance service 3. Baretich MF (2011) How to use financial benchmarks. Biomed
calls and key performance indicators. This in turn enables targeted Instrum Technol 45(5):405–407. https://doi.org/10.2345/0899-
service to be performed, which can be applied to the assessment 8205-45.5.405
4. Biomedical Engineering Advisory Group (BEAG) (2004) Life span
and modification of maintenance strategies and policies. In fact, of biomedical devices. Available: https://www.scribd.com/doc/
scheduled maintenance cannot always intercept problems before 29 48 713 11/B EAG-LifeS pan ofBi omed icalDev ices -
they arise. Maintenance processes must be continuously moni- GuidancePaper. Accessed 27 May 2019
tored, starting with the failures, so that the issues that most affect 5. Cohen T (2010) AAMI’s benchmarking solutions: analysis of cost
of service ratio and other metrics. Biomed Instrum Technol 44(4):
maintenance effectiveness can be identified and managed. 346–349. https://doi.org/10.2345/0899-8205-44.4.346
Data of the fault classification and KPIs were uploaded to the 6. Cohen T (2011) Staffing metrics: a case study. Biomed Instrum
business intelligence (BI) software QlikView® to obtain a global Technol 45(4):321–323. https://doi.org/10.2345/0899-8205-45.4.321
view of all equipment classes and results. Specifically, four types 7. CONSIP (2004) Disciplinare di gara a procedura aperta per la
prestazione dei servizi integrati per la gestione delle
of worksheets were conceived: one worksheet is related to main-
apparecchiature elettromedicali per le pubbliche amministrazioni.
tenance policy and types of faults, and the other is related to costs, In: Bandi di gara. Ministero dell’Economia e delle Finanze (I).
technological KPIs and corresponding maintenance times. A Available (in Italian): http://www.consip.it/bandi-di-gara/gare-e-
study is currently being carried out to test the use of most recent avvisi/gara-sigae-4. Accessed 27 May 2019
8. Dyro JF, Cheng M (2004) Good management practice for medical
BI open source software solutions, such as Apache Superset.
equipment. In: Dyro JF (ed) Clinical engineering handbook.
In this way, strategies can be modified so that an appropri- Elsevier Academic Press, San Diego, pp 108–110
ate compromise between the criticality of the equipment and 9. ECRI (2018) Institute website. Available: https://wwwecriorg/
the need to contain costs may be achieved. Pages/defaultaspx. Accessed 27 May 2019
10. Gonnelli V, Satta F, Frosini F, Iadanza E (2018) Evidence-based
Clinical engineering departments can promote failure reduc-
approach to medical equipment maintenance monitoring. IFMBE
tion by adopting strategies of training (to reduce errors related Proc 65:258–261. https://doi.org/10.1007/978-981-10-5122-7_65
to use and maintenance interventions with “no problem 11. Iadanza E, Dori F, Biffi Gentili G (2007) The role of bioengineer in
found”). They can also prepare tender specifications based up- hospital upkeep and development. IFMBE Proc 14:3641–3644.
on the evidence from these KPIs. Finally, they can design new https://doi.org/10.1007/978-3-540-36841-0_922
12. Iadanza E, Dori F, Biffi Gentili G, Calani G, Marini E, Sladoievich
protocols for internal maintenance based upon this analysis. E, Surace A (2007) A hospital structural and technological perfor-
A new management software, currently being developed at mance indicators set. IFMBE Proc 16:752–755. https://doi.org/10.
a regional level, is implementing the innovative set of KPIs 1007/978-3-540-73044-6_195
proposed in this work. Its commissioning will provide all the 13. Iadanza E, Marzi L, Dori F, Biffi Gentili G, Torricelli MC (2007)
Hospital health care offer. A monitoring multidisciplinar approach.
public hospitals in Tuscany with a great opportunity to apply IFMBE Proc 14:3685–3688. https://doi.org/10.1007/978-3-540-
the principles of evidence-based maintenance and will create 36841-0_933
the opportunity to compare different hospitals in future works. 14. Iadanza E, Baroncelli L, Manetti A, Dori F, Miniati R, Biffi Gentili
G (2011) An RFId smart container to perform drugs administration
reducing adverse drug events. IFMBE Proc 37:679–682. https://
Compliance with ethical standards doi.org/10.1007/978-3-642-23508-5_176
15. Italian Ministry of Health (2009) Recommendation number 9: pre-
Conflict of interest The authors declare that they have no conflict of vention of adverse events resulting in failure of medical devices or
interest. electromedical equipment. In: Ministero della salute. Available:
http://www.salute.gov.it/imgs/C_17_pubblicazioni_844_allegato.
Informed consent This article does not contain any studies with human pdf. Accessed 20 Jun 2019
participants or animals performed by the authors. 16. Luschi A, Marzi L, Miniati R, Iadanza E (2014) A custom decision-
support information system for structural and technological analysis
Open Access This article is distributed under the terms of the Creative in healthcare. IFMBE Proc 41:1350–1353. https://doi.org/10.1007/
Commons Attribution 4.0 International License (http:// 978-3-319-00846-2_334
 Med Biol Eng Comput

Valentina Gonnelli , B Sc. in

17. Miniati R, Dori F, Iadanza E, Fregonara MM, Biffi Gentili G (2011)
Electronics Engineering, M Sc.
Health technology management: a database analysis as support of
in Biomedical Engineering, win-
technology managers in hospitals. Technol Health Care 19(6):445–
ner of the HT Challenge Award
454. https://doi.org/10.3233/THC-2011-0642 in Management of Biomedical
18. UNI EN 13306:2010 Maintenance – Terminology Technologies, 18th National
19. UNI EN 15341:2007 Maintenance - Maintenance Key Performance Convention of Clinical Engineers
Indicators (AIIC) in 2018.
20. Wang B, Furst E, Cohen T (2006) Medical equipment management
strategies. Biomed Instrum Technol 40(3):233–237. https://doi.org/
10.2345/i0899-8205-40-3-233.1
21. Wang B, Fedele J, Pridgen B, Williams A, Rui T, Barnett L, Granade C,
Helfrich R, Stephenson B, Lesueur D, Huffman T, Wakefield JR,
Hertzler LW, Poplin B (2010) Evidence-based maintenance: part I-
measuring maintenance effectiveness with failure codes. J Clin Eng
35(3):132–144. https://doi.org/10.1097/JCE.0b013e3181e6231e
22. Wang B, Fedele J, Pridgen B, Williams A, Rui T, Barnett L, Granade C,
Helfrich R, Stephenson B, Lesueur D, Huffman T, Wakefield JR,
Hertzler LW, Poplin B (2010) Evidence-based maintenance part II-
comparing maintenance strategies using failures codes. J Clin Eng Francesca Satta , M Sc. in
35(4):223–230. https://doi.org/10.1097/JCE.0b013e3181f6b80b Computer Engineering, Clinical
23. Wang B, Fedele J, Pridgen B, Williams A, Rui T, Barnett L, Granade C, Engineer at the Careggi
Helfrich R, Stephenson B, Lesueur D, Huffman T, Wakefield JR, University Hospital and Meyer
Hertzler LW, Poplin B (2011) Evidence-based maintenance part III- Children’s Hospital in Florence
comparing maintenance strategies using failures codes. J Clin Eng (I), co-author of several articles
36(2):72–84. https://doi.org/10.1097/JCE.0b013e318214313c related to evidence-based mainte-
nance in both national and inter-
24. Wang B, Fedele J, Poplin B et al (2013) Evidence-based maintenance
national publications.
part IV-comparison of scheduled inspection procedures. J Clin Eng
38(3):108–116. https://doi.org/10.1097/JCE.0b013e31829a2a45
25. World Health Organization (2011) Medical equipment maintenance
programme overview. World Health Organization, Geneva
Available: http://www.who.int/medical_devices/publications/
medical-equipment-maintenance/en/ Accessed 27 May 2019
26. Ziken International (2005) How to manage series for healthcare
technology. Guide 2, how to plan and budget for your healthcare
technology. Available: http://apps.who.int/medicinedocs/
documents/s17710en/s17710en.pdf. Accessed 27 May 2019
27. Ziken International (2005) Guide 5. How to organize the mainte- Monica Gherardelli , Electronic
nance of your healthcare technology. In: ‘How to manage’ series for Engineer, PhD, Professor at the
healthcare technology. Sarah Townsend Ed. UK. Available: http:// University of Florence, scientific
apps.who.int/medicinedocs/documents/s17713en/s17713en.pdf. manager of agreements between
Accessed 27 May 2019 the Information Eng. Dept.
(University of Florence) and uni-
Publisher’s note Springer Nature remains neutral with regard to versity hospitals in Tuscany (I).
jurisdictional claims in published maps and institutional affiliations.

Ernesto Iadanza , Electronic

Engineer, BME, CE, MSc, PhD,
Adjunct Professor in Clinical
Engineering, University of
Florence. IEEE Senior Member.
Chairman of the IFMBE/Health
Technology Assessment Division
Board. Author of 145+ publica-
tions.