Spare Parts Management Methodology Development and Cost Evaluation - For offshore production facilities
Master thesis
Permanent lenke
http://hdl.handle.net/11250/2615021Utgivelsesdato
2017Metadata
Vis full innførselSamlinger
- Institutt for marin teknikk [3473]
Sammendrag
In the aftermath of the drop in oil price in 2014, the global energy industry has been in constant search for new business areas and more cost efficient solutions in order to stay competitive. The industry quickly recognized the cost-cutting within operation, but recently also the possibility of cost reduction within internal processes such as maintenance and spare parts management have been identified. Currently, there are no determined spare parts evaluation systems available, causing the companies to base their spare parts management on the recommendations from suppliers. This thesis is written in collaboration with Golar LNG and their FLNG vessel Hilli. Golar LNG have expressed their concern and presumption regarding excessive spare part recommendation, but have not found a suitable method to solve the problem.
Consequently, the thesis aims to develop a spare parts evaluation methodology that is suitable for offshore production facilities. In order to evaluate the effect and suitability of the methodology, it is implemented into two of the sub-systems on Hilli. Due to the need for reduced capital tie-up, a cost evaluation is performed after the implemententation.
The offshore industry is a risk prone industry, with large costs and downtime consequences. This is reflected in the international standards and regulations, where risk aware spare parts management systems are preferred. Based on these standards and relevant inventory theory, the developed methodology is presented in 5 steps:
1.Form the methodology foundation in order to clarify the needs and aims of the system2. Gather the relevant input data for further analysis e.g cost, lead time, spare part criticality and demand. 3.Classification of the spare parts utilizing multi-criteria analysis and AHP. The AHP evaluation criteria are developed based on the needs and aims identified in step 1.4. Analyze the spare parts based on their class. B and C-class components utilize EOQ, safety stock and reordering point, while A-class components are subjected to the newsvendor model and Poisson distribution. For uncertain input values, a sensitivity analysis is recommended. 5.Gather all the output data, present the result of the previous steps and report to the maintenance and risk department
The methodology was implemented into two of Hilli's sub-systems; (1) a refrigerant pump and (2) a flash gas offloading compressor. Golar LNG's aims and needs resulted in four evaluation criteria, where unit cost and criticality was equally important, annual dollar usage was essential and the lead time was the least important of the four.
Implementation of the methodology in the sub-systems on Hilli shows that the utilization of multi-criteria analysis tailors the classification to the companies needs and results in more uniform classes. This is an important result due to the analysis abilities to assure the critical components are subjected to the correct spare parts analysis addressing the issue of risk. Additionally, the methodology confirms Golar LNG's assumption of excessive spare parts recommendation, presenting an over-recommendation from the suppliers for 88\% of the spare parts. From this, we can draw the conclusion that the methodology is proved efficient in the areas of concern.
The cost was calculated both for the current spare parts inventory on Hilli and for the developed methodology. Due to uncertainties in regards to failure rate, demand and holding costs, a worst case and best case scenario was developed. When utilizing the worst case scenario in the cost evaluation, the results show a total reduction cost of approximately 140.000 USD for the refrigerant pump and 1.230.000 USD for the flash gas offloading compressor. In total, the cost reduction is 1.370.000 USD, a reduction of approximately 90\% from the current inventory cost. Consequently, the methodology is concluded to be cost efficient.
The results show a major cost effect and it is therefore recommended that a methodology based on mathematical and theoretical models as the one presented in this master thesis should be utilized over the recommendations of suppliers. Further work should be done to evaluate the availability effect more closely and to expand the methodology to more sub-systems on Hilli. Additionally, when Golar LNG's FLNG fleet is finished, an investigation of shared inventory should be performed.
The most critical uncertainty in this master thesis is the demand of the spare parts. Consequently, it is recommended that Golar LNG starts to track the inventory when Hilli is put in operation in order to establish a more accurate spare parts evaluation.