This research analyses the modes by which multiple risks could impact the performance of projects executed by the client company - a leading power generation company in India. It probes the case of load isolation experienced by one of the client's power generation units. It traces the instances to similar phenomenon observed in past research and analyses a wider spectrum of factors influencing the load mismatch of such a project. It also elaborates the multi- layered impact of factors including forecast of business decisions likely to be taken in the course of execution, as being done for IT projects. It generates possible scenarios that could be traced to the likely risk factors, based on analysis through established research models for project risk assessment. Based on this, it indicates the directions the client must investigate to construct a comprehensive risk framework for such projects.

Keywords

Power generation failures, Multi-layered risks, business risk mitigation, risk assessment models.

Introduction

The client company Tata Power Limited has a recently emerging concern in the operation of its Jamshedpur division. The frequent occurrence of load isolation of sale and non-sale bus islands due to system tripping is having an impact on its group operations, as it is a captive source for the Tata Steel unit and is also impacting its sale revenue vide export to the Damodar Valley Corporation (DVC) grid. This division is having total power generation capacity of 667.5 MW (Sale Island generating 240 MW and a Non-sale island with an estimated load of 530-560 MW)
Sale Island (bus) load is categorized as follows:

• Export to Damodar Valley Corporation (DVC) grid 50-60 MW
• Auxiliary power of 20 MW
• Tata Steel consumption of 140 MW (with 60 MW load pulsating type).

Non-Sale Island (Bus) load is categorized as follows:

• Four units of Jojebara (Units 1,2,4& 5) - Load of 370-380 MW
• Tata Steel captive power requirement of 100-110 MW.
• DVC import of 50-60 MW

Sale and Non-Sale Bus Isolation logic has been provided in the system to maintain Tata

• Steel - Tata Power system frequency when the lines from Tata Steel to DVC trip

For captive power generation plants like the Jamshedpur plant, such load isolation impacts the operations of the recipient industry. Captive power generation is the functional backbone of heavy industry units like Iron & Steel sector, chemicals and refineries, having a demand of more than 1 MW. It accounts for 19509 MW of installed power generation capacity in India (11.67% of total capacity, as per Central Electricity Authority, 2013)

Both power generation units and their customers, especially large organizations are increasingly becoming aware of the sensitivity of power failures and load isolation. Industries worldwide have begun to analyze the impact of intermittent power failure on their operations and facility, and capturing these risk scenarios in their Business Continuity Management strategies, as pointed by Bruch (2014).

The client company has identified a probable area where the root cause for the tripping could be lying, based on controlled experimentation in its grid, by installing a cut-off system in the logic. Its observations point that frequency variation of operating load is observed in instances of load isolation, indicating that there is a load-to-generated power mismatch in those instances.

The consultant group is able to deduce that beyond the technical design risks, strategic planning and decision risks Risk Identification in Power Generation Projects Due to Trip-Down Instances in Pre-Work Phase could also be involved in the performance of a system. In this case, the anticipation of future load increase in sale to the DVC grid and the housing expansion in Jojebara, in the planning phase, also has implications to the robustness of the system in responding to the load requirement. An analysis by this consultant group on the symptoms shows that this is not an isolated symptom –research by Barkans and Zalostiba (2007) identified a similar concern in generation units as a factor for black-outs. This is also echoed in the findings of Chandrakar et al (2012), which highlighted that impact of gradual increase in load demands for generation units in the load isolation phenomenon (islanding).

Context

The performance of a power generation unit is quite sensitive to the robustness of variation assessment done during the design phase (Om Prakash et al, 2011). For the design framework, this marks the importance of a pre-assessment of possible tripping modes that could occur. This would assist in predicting the failure instances. An example highlighting the importance of this concern would be that if this behavior is not predicted accurately, the system will have unplanned load isolation. When the sale bus is isolated, the immediate impact by load shedding is a cut of 50 MW which is part of the Export quota to the Damodar Valley Corporation (DVC) grid and a cut of 60 MW pulsating load due for the Tata Steel plant. When the non-sale bus is isolated, there is an immediate drop in the DVC import of 50-60 MW and Town load shedding of 40-50 MW, due to the load-cut of Jojobera units. As an aftermath, when one or more units get tripped, it hits a significant portion of the revenue generation of Tata Power business and its supply to the captive Tata Steel units.

The implications of failure risk identification in projects are not only limited to Tata Power application alone. Jaskowski and Biruk (2011) have highlighted that on a general framework, all capital projects like construction projects have multiple risk factors which can affect the execution or cost of the project. The research conducted by them also presented a framework to assess the risks and simulate the impact on performance in terms of instances of failure.

Recent trends

Project owners across sectors have identified the need to identify risks in the pre-work phase, so as to plan for mitigation against possible fallout. A lot of progress in risk assessment is observed in the software industry – Kop et al (2011) proposed a multi-pronged risk assessment model which was used to assess Technical, Executive and decision-making risks in an ERP implementation project. Brandas et al (2012) took this a step further by developing a general framework for IT development projects with multiple methodologies to analyze risk for a project. This has had successful improvement in new software projects from 2011, especially in SQL (Structured Query Language) application developers and Industrial Automation Systems, as identified by Biffl et al, 2011. Such a holistic risk assessment in the planning phase of these development projects has shown a consistent improvement in success rate of project execution.

Approach to problem / problem definition

It can be deducted from the symptomatic observations of the issue that the problem lies in the prediction of factors that could affect the future load demand for the project like strategic business decisions, expanding township loads, etc., in the design phase. This has helped narrow down the following models to approach the problem.

Cascaded approach to collective risk contribution

Research by Barkans and Zalostiba (2007) identified large over-loads of system cross-sections & tripping of lines, voltage avalanches, ground faults due to sagging of wires and frequency avalanches as the major causes of power system failure. Isolation logics were thus introduced into the system as a short term restoration against unloading of cross-sections. They proposed that an ideal self-restoration mechanism should facilitate auto-synchronization along with automatic re-closing of tripped lines. One of the viable approaches to construct a risk assessment map would be to take a cascaded model including the above stated multiple factors.

System behavior model

The lag in generator responsiveness to the change in load (as a shortcoming of the system design) was identified as another key cause for such system tripping by Om Prakash et al (2010). A model including multi-layered risk factors i.e., both at the design level and at the operating level, as proposed by Sanchez and Rios (2011) would prove to be effective.

Stochastic framework to identify combined impact

This model will proceed to identify the likely factors at play which might result in tripping of the system and load isolation. Such a framework could assess the combined impact of these multiple factors towards possible system failure through a stochastic model, similar to the one used by Adegun et al (2011). This would allow the organizations to identify and mitigate likely system failures during the design phase of such projects.

Utility distribution model

Research by Paul (2012) indicates that an urban planning model should accommodate for uniform distribution of services. Load sharing of utility services should follow equity load-shedding in the face of short-fall and system design must be modeled to ensure this.

Multi-layered business risk model

Business risks are to be factored in the planning phase including multiple layers like execution, strategic decision- making risks into the performance of a project (Kop et al, 2011). This is quite relevant for capital projects also as in this case, considering that the impact of business decisions of catering to expanding customer segments must be considered in the pre-plan phase of a project.

Contract outsourcing model for mitigation of business risk

An established method of project risk mitigation in IT projects, especially in public service providers is by segmental outsourcing. Research by Cox et al (2011) on IT outsourcing in public services by the local governments in UK explored 2 different approaches to outsourcing of contracts namely a sector wise division of utility provision and a scope-wise outsourcing approach for de-risking project load.

Considering the relevance of the above stated models in approaching the problem with respect to the application at hand, the consultant group opines that the following approaches are to be pursued further in addressing this problem:

1. Hybrid approach:Combination of the Multi-layered business risk model with the equity distribution model for utilities.
2. Client - Satellite agency model:Business risk mitigation model based on the Client - outsourced contract agency model.

The above 2 approaches will be probed in detail in relevance to the specific case faced by the client, for deducing the methodology to construct a risk framework for this application.

Literature review

The Hybrid approach (Approach 1) offers the advantage of planning for multiple future decisions that would influence the system from meeting demand while also identifying a structured mitigation process for a service shortfall. Academic research on the avalanche impact of multiple factors leading to execution issues in a project indicates the different combinations of factors could result in varying fallouts than that would result in isolation. The model developed by Adegun et al (2011) to analyze industrial accidents in oil and gas industries, due to combinations worker-related factors and their different impacts, offers a perspective into the system behavior that could be forecasted during the planning phase. The multi-layered approach proposed by Kop et al (2011) is significant considering that business opportunities are continuously expanding and there is a perpetual expansion of scope for such utility projects. Hence the impact of future strategic decisions also needs to be factored in risk mapping for such a project. The second part of the Hybrid approach explores the mitigation decision for a service shortfall. Utility disparity in various areas of a locality could cause a negative perception of the service provider among consumers as highlighted by Paul (2012). Research on public utility distribution models, considering growth in consumption and rise in population, by Yao et al (2012) points that a shortage factor for utility load is to be considered in the model. In this research work, they analyzed the water shortage in the public utility network of Beijing and proposed that in face of shortage, there is a need for industrial structure adjustment to balance the load on the distribution system, considering widespread disruption of consumers' daily life, if there is a shortfall in essential services.?iegis and Štreimikien???? (2005) predicted on similar lines that energy distribution patterns would be a key factor in sustainable national development indicators.

Taking a clue from the work of Kipyegen et al (2012), where the risk scenarios for a software development project were identified on different layers in the plan phase, the client can envision similar risk scenarios during the plan phase of such power projects by the Hybrid approach like the impact of future power sale deals during execution or expansion on the system success rate, unit islanding during service distribution on account of shortfall, etc.

The second approach namely the Client- Satellite agency model gives the advantage of mapping the risks of each division into mapping the risk of the entire system. Going beyond outsourcing of segmental contracts in utility services as a method of de-risking, Cox et al (2011) proposed sector wise division of utility provision in which the scope would be within the core competency i.e. Utility quota sharing for different wards amongst individual units to be factored in planning phase and a scope-wise approach for de-risking namely the main unit retaining the core competency and distributing services that provide competitive advantage or efficiency improvement to satellite units. Wan et al (2010) proposed that such a demarcation of scope between the main entity and satellite entities (similar to a Client- Contract agency model) in software projects would require such inter-unit transfer risks also to be captured in the risk framework of the whole project.

Parallel deductions from the above mentioned case of risk assessment in Public utility services in the London local government body (Cox et al, 2011) to the context of the client's problem can be made. It can be inferred that taking a Client-Satellite agency model for risk mapping would identify scenarios like likelihood of division-wise performance against rising demand and the extent of meeting demand for the Sales and Non-sale buses, when the load is demarcated to the respective divisions.

The secondary research indicates that the context of the underlying problem would require an investigation guided by a combination of the above stated frameworks is required for identifying the potential risks for a project. This would enable the client organization to construct a Risk framework that would be applicable for risk mitigation in the pre-work phase for similar projects.

Discussion

The consultant group has analyzed the issue faced by the client namely the load isolation of its sale and non-sale units during tripping instances of its division. The inference of the client based on symptoms of the occurrences namely frequency mismatches point towards a load-to-generated power mismatch. Secondary research by the consultant group indicated that load isolation of generation units could be due to a gradual rise in load demand for the unit (Chandrakar et al, 2011) and that it is not an isolated phenomenon of the client company. Jaskowski and Biruk (2011) recognized the impact of such performance risks of capital projects and proposed a framework for assessing the impact of these risks in the planning phase of construction projects.

This consultant group has analyzed the problem using a broader approach based on the research of Kop et al (2011) on IT projects, to follow a multi-layered approach considering the impact of strategic risks like potential business decisions during the execution of the project which could cause scope expansion, along with technical risks. This is done in conjunction with the Utility distribution model (Paul, 2012 and Yao et al, 2012) which traces impact on load shedding for utility distribution in the face of load shortfall. Another approach identified by the consultant group is a Client-satellite agency framework used by Cox et al, 2011 for business risk mitigation in Public utility services for the London local body government. An investigation using the above approaches highlight that the load mismatches (performance failures) could be caused by multiple risks which have to be captured in the pre-work phase of such a project.

Managerial implications

The consultant group opines that further investigation by the client, in the direction indicated above would indicate the multiple levels of risk that would impact the performance of the project in course of its execution. This will show the client multiple areas of impact of business decision driven risks on the project like scope expansion for saleable load to DVC, load to the Jojebara units. Thus the client would be able to trace and map such risks for projects and ultimately control them as was demonstrated by Biffl et al (2012) in Industrial Automation Systems projects and Kop et al (2011) in an ERP implementation project.

On the other hand, an investigation of the project progress in the perspective of the Client-Satellite agency model would highlight sub-division wise risks and their implications on the whole system in terms of load sharing and identify areas of control. This would help the client to isolate the risks that need to be mitigated from each sub-division of the project, similar to the approach used by Cox et al (2011).

Research limitations/scope for future research

The consultant group would like to acknowledge that only 2 approaches which were identified closest to the client's project model. Also, the scope of this consultancy project is only the definition of the problem faced by the client.

Hence the required solution for the client, namely the risk mapping framework, (for use in planning of such power projects) will have to be traced by the client by further investigation using one of the above suggested models. Hence further research must be focused on choosing the most suited approach and investigating the risks that would have implications on the performance of the project. This would help the organization construct a generic risk framework for similar projects.

Acknowledgements

The project was made under the group of students; Aayush Goyal, Deepak Regunath, Hitesh Gomber, Maneesha Akshay, Rahul Sarda, SP Jain Institute of Management and Research, Mumbai, India who would like to acknowledge the valuable insights from Dr. Arup Mazumdar, Faculty in Charge, Business Consulting Projects Group, SPJIMR,. They would also like to thank Ms. Vibha Bhilawadikar, Vice-President, Kiesqaure Consulting, who mentored the group in this project.

References

1. Chandrashekhar Chandrakar, Bharti Dewani, Deepali Chandrakar (2012), “An assessment of distributed generation islanding detection methods”, International Journal of Advances in Engineering & Technology, Vol. 5, Issue 1, pp. 218-226.
2. Claudiu Brandas, Otniel Didraga, Nicolae Bibu (2012), “Study on Risk Approaches in Software Development Projects”, Informatica Economica (ISSN-1453-1305), Vol. 16 Issue 3, p148-157. 10p.
3. K. Adegun, H. A. Ajimotokan* & G. O. Oyelohunnu (2011), “The development of interactive software for assessing risks and estimating industrial man-hour loss, International Journal of Research & Reviews in Applied Sciences, Vol. 7 Issue 1, p48-53. 6p.
4. J. Barkans and D. Žalostiba, (2007), “Protection of generating units at deviations of parameters by sectioning a power system as a tool for blackout prevention”, Journal of Power & Electrical Engineering, 2007, Issue 20, p25-35. 11p.
5. Jiangping Wan, Dan Wan, Hui Zhang (2010), “Case Study on Business Risk Management for Software Outsourcing Service Provider with ISM”, Journal of Technology and Investment, 2010, 1, 257-266.
6. John A. Sanchez, Mario A. Ríos (2011), “Vulnerability Assessment of Cascade Contingencies in Power Systems”, International Review of Electrical Engineering (I.R.E.E.), Vol. 6, N. 4, p 1831-1839.
7. Michael Bruch and Larry Hunter (2014), “Power trips - Energy Risks - the dangers of power cuts and blackouts”, Expert Risk Article, Allianz Global Corporate and Speciality, January 2014.
8. Michael Cox, Martyn Roberts and John Walton (2011), “IT Outsourcing in the Public Sector: Experiences from Local Government”, Electronic Journal Information Systems Evaluation Volume 14 Issue 2.
9. NoelaJemutai Kipyegen, Waweru Mwangi and Stephen Kimani, “Risk Management Adoption Framework for Software Projects: A Case Study for Kenyan Software Project Managers and Developers”, International Journal of Computer Science Issues, Vol. 9, Issue 3, No 3, 365-373.
10. Om Prakash, T. Ravi Kiran and Ajay Kumar Kaviti (2010) “Artificial Neural Network based Prediction Model for reduction of failure frequency in Thermal Power Plants”, Archives of Applied Science Research, 2010, 2 (6): 98-107.
11. Om Prakash, Anil Kumar, T. Ravi Kiran and Ajay Kumar Kaviti (2011), “Anfis Based Prediction Model for Reduction of Failure Frequency in Captive Power Plant”, Archives of Applied Science Research, 2011, 3 (1): 52-64.
12. P. Jaskowski, S. Biruk (2011), “The conceptual framework for construction project risk assessment”, Journal of Reliability: Theory & Applications, Vol. 6 Issue 3, p27-35. 9p.
13. Remigijus ?iegis, Dalia Štreimikiene (2005), “Integration of Sustainable Development Indicators into Sustainable Development Programmes”, Journal of Economics of engineering decisions, Vol No 2 (42).
14. Suman Paul (2012), “Factors and Dimensions of Inter-Ward Disparities in Urban Facility Utility Services in Burdwan City, India”, Archives of Applied Science Research, 4 (3):1376-1388.
15. Ye?im Kop, H. Ziya Ulukan and Tuncay Gürbüz (2011), “Evaluating the Failure Risk Level of an Enterprise Resource Planning Project using Analytic Network Process in Fuzzy Environment”, Journal of Multiple-Valued Logic & Soft Computing, Vol. 17 Issue 4, p407-423. 17p.
16. Yi-qun Yao, Yuan-biao Zhang, Jin-ding Liu, Zhong-jie Shen, Bo-wen Liu (2012), “Model for Evaluating Urban Water Shortage Risk: A Case Study in Beijing”, International Journal of Digital Content Technology & its Applications, Vol. 6 Issue 6, p68-79. 12p.

Tags: MET Institute of Management