SUPPLY CHAIN INNOVATION AGILITY, FIRM SIZE AND DEMAND FORECASTING PERFORMANCE OF OIL MARKETING COMPANIES IN KENYA

Abstract

Demand forecasting performance has remained a challenging subject in supply chain management due to the difficulty involved in accurately planning, tracking, and monitoring demand in unpredictable markets. Innovation plays a critical role in improving forecasting accuracy and enabling real- time demand tracking. The purpose of this research was to determine the significance of actively incorporating agile supply chain innovations in oil marketing companies to enhance the accuracy and timeliness of demand forecasting. Specifically, the study sought to analyse the influence of business process innovation on the demand forecasting performance of oil marketing companies in Kenya; examine the influence of technological innovation on the demand forecasting performance of these companies; assess the influence of network structure innovation on demand forecasting performance of oil marketing companies in Kenya; establish the impact of customer- centric innovation on demand forecasting performance of oil marketing companies in Kenya; and evaluate the moderating influence of firm size moderates the relationship between supply chain innovation agility and demand forecasting performance. The research was guided by four theories: Dynamic Capabilities Theory, Systems Theory, Resource- Based Theory, and Network Theory. A descriptive design with a mixed- methods approach was adopted. The target population comprised 145 registered oil marketing companies in Kenya. A purposive sampling technique was used to select 81 companies. The study utilised both open- ended questionnaires and secondary data sources. Quantitative data were analysed using SPSS version 27, while qualitative data were analysed manually. The adjusted R2indicated that 81.1% of the variables accounted for demand forecasting performance, with BPI and TI showing positive but not significant effects, and CCI and NSI demonstrating positive and significant effects. Introducing firm size as a moderator increased the adjusted R2to 94.5%. However, after moderation, there was a partial significant effect, and the influence on demand forecasting performance decreased; BPI (β = -0.065, p = 0.018), NSI (β = -0. 078, p = 0.273), CCI (β =- 0.177, p = 0.005) had negative effects, while TI (β = 0.047, p = 0.513) had a positive effect. The study concluded that, individually, the SCIA components were positive and significant; however, when regressed together, they tend to share resources, thereby reducing explanatory power and significance. The study also concluded that firm size moderation was partial and context-specific. The study thus recommends decentralising decision-making, empowering regional units, streamlining internal processes to increase flexibility and responsiveness, and investing in integrated digital platforms, real-time data-sharing systems, and collaborative forecasting practices to improve visibility across the supply chain. It also proposes leveraging advanced customer analytics, market intelligence systems, decentralised customer relationship management, and IoT and digital twin technologies that precisely adjust demand, monitor petroleum consumption rates, and visualise the supply chain to identify trends that may influence future demand.