Summary of the Project
By combining commonly, yet disparately, used reliability and maintenance engineering techniques in a structured yet agile manner, SAS Asset Management have enabled NSW Health in better achieving an appropriate balance between cost, risk and performance.
Whilst recognising and accepting challenges such as limited to zero configuration or failure data, the team at SAS-AM were able to establish and combine enough data from subject matter experts along with their associated levels of uncertainty to produce NSW Health’s first data-driven asset lifecycle cost data model whilst enabling future improvement and maturity increase.
Description of the Project and Framework Developed
The Illawarra-Shoalhaven Local Health District (the LHD) identified the need to better support their asset management budgetary applications through the production of a data-driven and defendable Life Cycle Cost Model (the Model) for their hospital beds asset class.
The Model, in order to be considered defendable, was required to couple the financial requests with its expected resultant outcomes:
- The expected level of service (performance);
- The resultant residual risk; and,
- An uplift in technical data.
SAS Asset Management (SAS-AM) deployed its Asset Dependability Assurance Framework (ADAF) in pursuit of the LHD’s three required outcomes.
SAS-AM’s ADAF combines a number of typically disparate maintenance and reliability engineering techniques with asset management context and insight to produce tangible asset management outcomes whilst enabling:
- An agile delivery approach;
- A logical and systematic suite of steps ;
- The organisation to learn about itself and its asset portfolio;
- An understanding of the current asset performance capability; and,
- A continuous improvement cycle for future iterations.
The technical data for the hospital beds asset class was not available within the LHD’s systems. This included:
- Asset configuration data;
- Population data;
- Failure or event data;
- Required performance targets or historical data; or,
- Fleet-level operational performance targets.
SAS-AM’s ADAF comprises of eight discreet elements which are designed to enable its clients to overcome challenges such as those faced by the LHD. The following sections outline the outcomes of each of the ADAF elements and their value-add to the LHD.
Asset Condition Review – ACR
A series of workshops with the LHD Subject Matter Experts (SMEs) were designed to extract implicit knowledge and data from the LHD SMEs.
As a result, the LHD had developed its own understanding of the above. Some examples are provided below.
|Configuration Type 1
|Configuration Type 2
|Configuration Type 3
|Configuration Type 4
|General Hospital Beds
|Bariatric Hospital Beds
|Bulli Hospital & Aged Care Centre*
|David Berry Hospital
|Port Kembla Hospital
Table: Asset Population Data, per configuration type, per hospital
Figure: Generic System Breakdown Structure
Figure: General Hospital Beds System Breakdown Structure
Figure: Bariatric Hospital Beds System Breakdown Structure
Figure: Birthing Hospital Beds System Breakdown Structure
Figure: ICU/HDU Hospital Beds System Breakdown Structure
Failure Modes Analysis – FMECA
The FMECA was undertaken with the LHD SMEs to establish the known risk profile of each of the configuration types and their subsystems.
The result was both an understanding of failure mode risk (O x S) as well as the risk priority number (O x D x D) for each subsystem, for each hospital bed and for each hospital.
Figure: Hospital Beds Failure Mode Risk Profile
Figure: Hospital Beds Risk Priority Number Profile
Baseline Reliability Analysis– BRA
The BRA utilised the outputs of the ACA and the FMECA occurrence data and produced a series of reliability block diagrams.
These diagrams enabled an analytical model of which each configuration type’s reliability for a nominal 1-year period was derived.
The outcome for the LHD was a thorough understanding of the reliability capability of their assets.
Figure: ICU/HDU Hospital Beds Baseline Reliability
Figure: Birthing Hospital Beds Baseline Reliability
Figure: Bariatric Hospital Beds Baseline Reliability
Figure: General Hospital Beds Baseline Reliability
Maintenance Plan Development – MPD
The MPD utilised the Reliability Centred Maintenance 3 (RCM3) framework to identify the most appropriate maintenance plan on a per hospital bed basis – without consideration for the ‘fleet’ aspect of the broader system.
The analysis identified that there should be 17x in situ inspections tasks, 4x preventative tasks and 47x no-scheduled-maintenance (run to failure) tasks.
Availability Capability Assessment – ACA
The ACA utilised the failure profiles and diagrams from the BRA and simulated these in concert with the single hospital bed maintenance plan developed as part of the MPD.
The monte carlo simulation identified that several of the RCM3 tasks were insufficient to achieve the fleet-level performance targets.
As a result, the single hospital bed maintenance plans were tailored to ensure the fleet could achieve its required capability.
Figure: Shellharbour Hospital Beds Availability Capability
Figure: Shoalhaven and Coledale Hospital Beds Availability Capability
Figure: Port Kembla Hospital Beds Availability Capability
Figure: David Berry Hospital Beds Availability Capability
Figure: Milton-Ulladulla Hospital Beds Availability Capability
Maintenance Task Determination & Maintenance Task Strategy – MTD & MTS
The MTD defined the resource requirements of the resultant maintenance plan derived from the ACA whilst the MTS defined the most appropriate procurement model (insource vs outsource).
As a result, the LHD had a comprehensive understanding of their asset require requirements.
Lifecycle Cost Analysis – LCA
The final element of the ADAF combined the analysis into a monte carlo simulation to determine the resultant asset cost, performance and risk profile across the predefined 8-year analysis period.
The LCA produced an annual forecast (complete with uncertainty confidence bounds) of resource requirements for each hospital bed fleet.
The outcome enabling the LHD to understand the balance between cost, risk and performance for their hospital bed asset class.
Contribution made to the LHD.
As a result of our work, the team at the LHD now have an improved understanding of their asset portfolio, its requirements, capability, and cost profile, specifically a better understanding of:
- What the hospital beds asset class portfolio comprises e.g. the asset populations on a per hospital, per configuration type basis, the system-subsystem-interfaces present within each configuration type, commonalities between configuration types and component population data;
- The perceived asset risks presented by the configuration types as well as an understanding of criticality/more-less risky subsystems enabling future prioritisation;
- The baseline reliability capability of each hospital, configuration type, system and/or subsystem within their portfolio for a nominal 1-year period. This understanding was developed in concert with SME expertise and asset configuration data;
- An optimised and consistent fleet-wide maintenance plan for each configuration type tailored to the specific fleet compositions and considerations. This builds on the standard RCM3 framework to optimise the maintenance plan in the context of fleet assets (redundancy etc.).
- The resource requirements derived from the optimised fleet-wide maintenance plan monte carlo simulations. This included a thorough understanding spare parts (both logistic delays and store quantity), labour resource (both corrective, preventative and condition monitoring) requirements, special tooling, and outage requirements for the given 8-year period; and
- The 8-year lifecycle cost requirements of the asset class on a per hospital, per configuration type basis.
SAS-AM’s ADAF has enabled NSW Health to adequate plan for the next 8-year period with a thorough understanding of the cost, risk and performance outcomes of its asset base. Furthermore, the LHD team can undertake trade off analyses to better understand the resultant performance and risk profiles for a given funding envelope.
SAS-AM and the LHD recognise that, due to the incomplete and missing input data, the Model consists of a significant amount of uncertainty however this is actively communicated.
Furthermore, both parties recognise the value in simply utilising the ADAF to better understand the context of the asset class and the creation of a data model for future improvement through additional iterations.
It is expected that, with the improved understanding of the context of the hospital beds asset class throughout the LHD, conditions and level of service through the hospital network will improve as well as the fiscal performance of the LHD.