**2. The LCPA Tool**

The acronym LCPA appeared for the first time some years ago on the scenario of EU funded project (BEST, JOULES, RAMSES, and more recently SHIPLYS) since a fully compliant LCA procedure is simply not applicable to ships because of their complexity.

In HOLISHIP, evolution has been proposed, with a specific focus on the possibility to obtain a single and comprehensive index to characterize the ship performance, inclusive of both economic and environmental terms.

The LCPA tool is an integrated software to allow for a life-cycle ships assessment during the design phase. LCPA efficiently merges the LCC and LCA performances that, therefore, can be observed comprehensively from the perspective of the designer/ship-builder, as well as of the ship-owner. The calculation of selected Key Performance Indicators (KPIs) enables the evaluation of the LCPA Index that describes the performance of a baseline ship compared to other alternative design configurations [7,9]. The overall index is a linear combination of selected KPIs that have been properly normalized. The relevant formulations are reported in the following Equations (1)–(3), but further details can be found in [7,10]. In principle, the methodology is open for the integration of further aspects, besides economic and environmental ones, e.g., safety performances [11], in case adequate KPIs are identified.

$$I\_{\rm LCC} = \sum\_{i=1}^{N\_{\rm LCC}} f\_{i,\rm LCC} \ast \mathbf{c}\_{i,\rm LCC} \le 1 \; ; \; where \; : \; \sum\_{i=1}^{N\_{\rm LCC}} f\_{i,\rm LCC} = 1 \tag{1}$$

$$I\_{LCA} = \sum\_{i=1}^{N\_{LCA}} f\_{i,LCA} \* c\_{i,LCA} \le 1 \; ; \; where \; : \; \sum\_{i=1}^{N\_{LCA}} f\_{i,LCA} = 1 \tag{2}$$

$$I\_{\rm LCPA} = f\_{\rm LCC} \* I\_{\rm LCC} + f\_{\rm LCA} \* I\_{\rm LCA}; \text{ where } f\_{\rm LCC} + f\_{\rm LCA} = 1 \tag{3}$$

It is worthwhile mentioning that different approaches are possible to combine economic and environmental aspects, for example, converting all incomparable values into monetary values [12]. To better understand the LCPA software and its further improvements developed in this paper, a synthesis of the main topics are proposed in the following sections. As already mentioned, Key Performance Indicators (KPIs) are the base of the LCPA tool. They are performance indicators resulting after the definition of the problem and its cost drivers. There are, respectively, economics and environmental KPIs that play a fundamental role in the decision-making process:


Further considerations about the significance of such KPIs can be found in [18]. explicit equations for estimation of all these values, are given in [7,10].

The relations between these KPIs are provided in Figure 1a–c. It is important to denote that the CAPEX value has been described in the paper for the sake of completeness, but during this work, we have always referred to the BLD value.

**Figure 1.** (**a**) CAPEX (Capital Expenditure) explanation; (**b**) OPEX (Operating Expenditure) explanation; P&I stands for Protection & Indemnity; MCR is the Maximum Continuous Rating; (**c**) NPV (Net Present Value) explanation. BLD: Building Cost; M&R: Maintenance and Repair costs.

A thorough explanation of the LCPA process is given in [7,10]. Nevertheless, in Figure 2, a block diagram of LCPA steps is provided. The same steps have been followed in the organization of the application case below.

**Figure 2.** Block diagram illustrating LCPA (Life Cycle Performance Assessment) steps. KPIs: Key Performance Indicators; EEDI: Energy Efficiency Design Index.

In this work, the M&R costs evaluation have been improved to permit a better evaluation of the OPEX costs: the empiric formulation used in the current first version of LCPA tool could be replaced with a flexible prediction model based on real maintenance actions. The maintenance model, discussed in paragraph 4, would be implemented as part of the LCPA tool. A different energy system layout might also affect BLD value. In this work, such an issue was taken into account.

#### **3. Maintenance Strategies**

Maintenance actions ensure that a system performs in the best way during its whole life cycle, preserving its integrity and performances over time. Different maintenance techniques were developed in the last decades to better preserve system capabilities during its life cycle, minimizing the failure rate and downtime. All these actions can be summarized as follows [19].

**Corrective maintenance**: the simple one without any scheduled action. The operator attends when a failure occurs in switching off the system and performing the maintenance with more or less important economic implications. This approach is based on the belief that the costs sustained for downtime and repairs, in case of a fault, are lower than the investment required for a maintenance program. This strategy may be cost-effective until catastrophic faults occur.

**Scheduled preventive maintenance**: a step forward in maintenance policy. The manufacturer provides a so-called Mean Time Between Maintenance (MTBM) that is the best working time range when a maintenance action has to be performed to prevent system failure or degradation. In this way, an operator can plan the maintenance services to minimize the impact on working hours and so on costs and profits. The maintenance cycles are planned according to the need to take the device out of service. The incidence of operating failures is reduced. In a complex system with more sub-systems that work together to complete a task, this method can be a better way to plan the maintenance operations.

**Performance-based maintenance**: often indicated as Condition Based Maintenance (CBM), it is based on the response analysis of multiple sensors mounted on the system to measure actual working parameters like temperature, pressure, fluid levels, and more. Through prediction models, they are automatically compared with average values and performance indexes. Maintenance is carried out when some indicators give the signaling that the equipment is deteriorating and the failure likelihood is increasing. This strategy, in the long term, allows a drastic reduction in maintenance costs, thereby minimizing the occurrence of serious faults. With these previsions, the operator can plan one or more maintenance action only when it is requested and avoiding the interventions when it is not necessary.

These three actions are the most common, but other two maintenance policies can be further mentioned, namely the **adaptive maintenance** and the **perfective maintenance**. The first one is applied when the system needs to evolve and adapt for new needs or working contest to extend the system working life; the second one is performed when there is the chance to improve a system with the installation of innovative technology.

In this work, the scheduled preventive maintenance was assumed to develop the maintenance prediction model where the designer could evaluate the quality of the different configuration in relation to maintenance costs and time. This model would improve the current version implemented in the LCPA tool, based on an empirical formulation of the maintenance costs. Preventive maintenance is defined as a semi-deterministic model; however, it gives an effective way to compare different layouts and operational profiles at an early design stage: it takes into account, in fact, that designers could not have access to a large database, recorded dataset, or detailed information on systems they are investigating. The main information needed to set up a scheduled maintenance approach could be more easily obtained from manufacturer manuals. This is the reason why we used the preventive maintenance technique as a starting point [20,21]. The corrective maintenance and the dry-dock maintenance costs have been disregarded in this paper, and they are going to be taken into consideration later in the research activity.
